JP2022163058A - Stereo signal coding method and stereo signal encoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a stereo signal coding method and a stereo signal encoder that reduce high-frequency distortion of decoded stereo signals as much as possible while improving the spatial sense and sound image stability of the decoded stereo signals.

SOLUTION: A coding method has a step 301 of determining residual signal coding parameters for a current frame of a stereo signal based on downmix signal energy and residual signal energy of each of M subbands of the current frame. The residual signal coding parameters of the current frame are used for indicating whether to code the residual signal of the M subbands. The M subbands are at least a portion of N subbands, where N is a positive integer greater than 1, M≤N and M is a positive integer. The method further includes a step 302 of determining whether to encode the residual signal of the M subbands of the current frame based on the residual signal coding parameters of the current frame.

SELECTED DRAWING: Figure 6

COPYRIGHT: (C)2023,JPO&INPIT

Description

This application is based on Chinese Patent Application No. 201810549237.3 entitled "STEREO SIGNAL ENCODING METHOD AND APPARATUS" filed with the Chinese Patent Office on May 31, 2018, which is incorporated herein by reference in its entirety. priority based on

The present application relates to the audio field, and more particularly to a stereo signal encoding method and stereo signal encoding apparatus.

The general process of encoding a stereo signal using time domain or time frequency domain stereo encoding techniques is as follows.
performing time-domain preprocessing on the left-channel time-domain signal and the right-channel time-domain signal;
performing time domain analysis on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing;
performing time frequency domain transformation on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing to obtain a left channel frequency domain signal and a right channel frequency domain signal;
determining the Inter-channel Time Difference (ITD) parameter in the time domain;
performing time shift adjustments to the left frequency domain signal and the right channel frequency domain signal based on the ITD parameters;
Compute stereo parameters, a downmix signal and a residual signal based on the left channel frequency domain signal and the right channel frequency domain signal obtained by time shift adjustment, and encode the stereo parameters, the downmix signal and the residual signal. .

In the prior art, only the stereo parameters and the downmix signal are generally coded when the coding rate is relatively low, and part or all of the residual signal is coded only when the coding rate is relatively high. It has been known. In this case, the spatial sensation of the decoded stereo signal is relatively low, and the sound image stability of the decoded stereo signal is relatively low.

In other prior art, it is known that when the coding rate is relatively low, in addition to the downmix signal, residual signals of subbands satisfying a preset bandwidth range are also coded. Although this coding method can improve the spatial perception and sound image stability of the decoded stereo signal, the total number of coding bits used for coding the residual signal and coding the downmix signal is Fixed, the low frequency information is preferentially encoded during downmix signal encoding, so that when the downmix signal is to be encoded, some signals are replaced with more abundant high frequencies in the downmix signal. There may not be enough bits to encode the information. Therefore, the high frequency distortion of the decoded stereo signal is relatively large, which affects the coding quality.

The present application aims to improve the spatial sensation and sound image stability of the decoded stereo signal and to reduce the high frequency distortion of the decoded stereo signal as much as possible, thereby improving the coding quality. , to provide a stereo signal encoding method.

According to a first aspect, a stereo signal encoding method is provided. The method comprises determining residual signal coding parameters for a current frame of a stereo signal based on downmix signal energy and residual signal energy for each of M subbands of the current frame, comprising: The residual signal coding parameter of the current frame is used to indicate whether to code the residual signal of the M subbands, where the M subbands are at least part of the N subbands. , where N is a positive integer greater than 1, M ≤ N, M is a positive integer, and the M subs of the current frame based on the residual signal coding parameters of the current frame and determining whether to encode the band's residual signal.

The residual signal coding parameters are determined based on the downmix signal energy and the residual signal energy of the M subbands within the N subbands, satisfying a preset bandwidth range; Whether to code the residual signal of each subband is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, the M subbands are the preset maximum subband index number among the N subbands. Here are the M subbands:

Optionally, in one embodiment, the M subbands have a subband index number greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number in the N subbands. are M subbands.

The minimum subband index number and/or maximum subband index number are set based on different coding rates. The residual signal coding parameters are determined based on the different coding rates and the downmix signal energy and residual signal energy of multiple specific subbands within the N subbands, and Whether to code each residual signal is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, the residual signal of each of the M subbands is encoded based on the residual signal coding parameters of the current frame. The step of determining whether the residual signal coding parameter of the current frame is compared to a preset first threshold, wherein the first threshold is greater than 0 and less than 1.0, the step and determining not to encode the residual signal of each of the M subbands if the residual signal coding parameter of the current frame is less than or equal to the first threshold; or and determining to encode the residual signal of each of the M subbands if the encoding parameter is greater than a first threshold.

A first threshold is set and the determined residual signal coding parameter is compared to the first threshold. Whether to encode the residual signal of each of the M subbands is determined based on the result of comparing the residual signal encoding parameter with the first threshold. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to encode all residual signals in subbands that satisfy a preset bandwidth range is determined based on the result of comparing the residual signal coding parameter with the first threshold. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, residual signal encoding of the current frame based on downmix signal energy and residual signal energy for each of the M subbands Determining the parameters includes determining residual signal coding parameters based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands.

A residual signal coding parameter for each of the M subbands is determined based on the downmix signal energy, the residual signal energy, and the side gain to encode the residual signal for each of the M subbands. is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, residual signal encoding based on downmix signal energy, residual signal energy, and side gain for each of the M subbands Determining the parameters includes determining a first parameter based on the downmix signal energy, residual signal energy, and side gain for each of the M subbands, wherein the first parameter is M second step based on the downmix signal energy and the residual signal energy for each of the M subbands, indicating a value relationship between the downmix signal energy and the residual signal energy for each of the M subbands; wherein the second parameter indicates a value relationship between the first energy sum and the second energy sum, and the first energy sum is the residual signal of the M subbands a sum of the energy and the downmix signal energy, a second energy sum being the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency-domain signal of the previous frame of the current frame; a step in which the M subbands of the current frame have the same subband index numbers as the M subbands of the previous frame; determining residual signal coding parameters for the current frame based on the long-term smoothing parameters.

Referring to the first aspect, in one possible implementation of the first aspect, the first parameter is determined based on the downmix signal energy, the residual signal energy, and the side gain of each of the M subbands. The determining step is determining M energy parameters based on the downmix signal energy, residual signal energy, and side gain for each of the M subbands, wherein the M energy parameters are equal to M step, respectively showing a value relationship between the downmix signal energy and the residual signal energy of each of the subbands, the M energy parameters corresponding one-to-one with the M subbands; and determining as the first parameter the energy parameter having the maximum value among the energy parameters of .

Referring to the first aspect, in one possible implementation of the first aspect, the energy parameter of the subband with subband index number b among the M energy parameters satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/(res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g(b) is 4 represents a function of the side gain side_gain[b] of the subband whose subband index number is b.

Referring to the first aspect, in one possible implementation of the first aspect, residual signal encoding of the current frame based on downmix signal energy and residual signal energy for each of the M subbands Determining the parameter is determining a first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being the and determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands. wherein the second parameter indicates a value relationship between the first energy sum and the second energy sum, the first energy sum being the residual signal energy of the M subbands and the downmix is the sum of the signal energies, the second energy sum is the sum of the residual signal energies and the downmix signal energies of the M subbands in the frequency domain signal of the frame before the current frame, and the sum of the downmix signal energies of the current frame; A step in which the M subbands have the same subband index number as the M subbands of the previous frame, and the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame. determining the residual signal coding parameters for the current frame based on .

Referring to the first aspect, in one possible implementation of the first aspect, determining the first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands comprises: , determining M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, wherein the M energy parameters are based on the downmix signal energy and the residual signal energy for each of the M subbands. Steps and maximum values among the M energy parameters, each showing a value relationship between the mix signal energy and the residual signal energy, wherein the M energy parameters correspond one-to-one with the M subbands; as the first parameter.

Optionally, in one implementation, the sum of the M energy parameters is determined as the first parameter res_dmx_ratio ₁ (to be corrected), where res_dmx_ratio ₁ is the maximum of the M energy parameters res_dmx_ratio_max and Corrected based on the downmix signal energy res_cod_NRG_M[b] of each of the M subbands to determine res_dmx_ratio ₂ obtained by correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, in one implementation, the res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

Optionally, in one embodiment, the encoder side determines the first parameter based on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands.

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, in one implementation, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the following equation:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, in one implementation, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the following equation:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr

Referring to the first aspect, in one possible implementation of the first aspect, the energy parameter of the subband with subband index number b among the M energy parameters satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] res_cod_NRG_M[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b.

Referring to the first aspect, in one possible implementation of the first aspect, the current frame residual signal coding parameter is the current frame long-term smoothing parameter, and the current frame long-term smoothing parameter parameter satisfies the following equation,
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is the same as the preset second is greater than the value of α for greater than or equal to the threshold of , the second threshold is between 0 and 0.6 inclusive, the third threshold is between 2.7 and 3.7 inclusive, or the second parameter is prior to The value of α when the first parameter is greater than the preset fourth threshold is less than the preset fifth threshold, and the value of α when the first parameter is less than or equal to the preset fourth threshold. greater than the value of α and the fourth threshold is between 0 and 0.9 and the fifth threshold is between 0 and 0.71 or the second parameter is greater than or equal to the preset fifth threshold and is less than or equal to a preset third threshold, then the value of α is less than the first parameter preset second threshold and the second parameter is preset third threshold less than the value of α if greater than, the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0.71 It is below.

Referring to the first aspect, in one possible implementation of the first aspect, when it is determined to encode the residual signal of M subbands, the method comprises: encoding the downmix signal and the residual signal, or encoding the downmix signal for the M subbands when it is determined not to encode the residual signal for the M subbands; Including further.

According to a second aspect, an encoding device is provided. The apparatus is configured to determine residual signal coding parameters for a current frame of the stereo signal based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame. A first determining module, wherein the residual signal coding parameters of the current frame are used to indicate whether to code the residual signal of M subbands, wherein the M subbands are N a first determining module, wherein N is a positive integer greater than 1 and M≤N, where M is a positive integer; and residual signal encoding of the current frame. and a second decision module configured to decide whether to encode the M subband residual signals based on the parameters.

According to a third aspect, an encoding device is provided. The apparatus includes a memory and a processor, the memory configured to store the program, the processor configured to execute the program, and when the program is executed, the processor performs the first aspect or the first aspect. A method according to any one of the possible embodiments of aspect 1 is carried out.

According to a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores program code to be executed by the device, the program code including instructions used to perform the method according to the first aspect or various implementations of the first aspect.

According to a fifth aspect, a chip is provided. The chip includes a processor and communication interface. A communication interface is configured to communicate with an external device. The processor is configured to perform the method according to the first aspect or any one of the possible implementations of the first aspect.

Optionally, in one implementation, the chip may further include memory. The memory stores instructions and the processor is configured to execute the instructions stored in the memory. When the instructions are executed, the processor is configured to perform the method according to the first aspect or any one of the possible implementations of the first aspect.

Optionally, in one embodiment, the chip is incorporated into terminal equipment or network equipment.

1 is a schematic structural diagram of stereo encoding and decoding in the time domain according to an embodiment of the present application; FIG. 1 is a schematic diagram of a mobile terminal according to an embodiment of the present application; FIG. 1 is a schematic diagram of a network element according to an embodiment of the application; FIG. 1 is a schematic flow diagram of a stereo signal encoding method in the frequency domain; 1 is a schematic flow diagram of a stereo signal encoding method in the time-frequency domain; 1 is a schematic flow diagram of a stereo signal encoding method according to an embodiment of the present application; Fig. 4 is another schematic flow diagram of a stereo signal encoding method according to an embodiment of the present application; 1 is a schematic block diagram of a stereo signal encoding device according to an embodiment of the present application; FIG. Fig. 3 is another schematic block diagram of a stereo signal encoding device according to an embodiment of the present application;

The technical solutions of the present application are described below with reference to the accompanying drawings.

FIG. 1 is a schematic structural diagram of a stereo encoding and decoding system in the time domain according to an exemplary embodiment of the present application. Stereo encoding and decoding system includes encoding component 110 and decoding component 120 .

Encoding component 110 is configured to encode the stereo signal in the time domain. Optionally, encoding component 110 may be implemented using software, or may be implemented using hardware, or may be implemented in the form of a combination of software and hardware. . This is not limited in this embodiment.

Encoding component 110 encodes the stereo signal in the time domain and includes the following steps.

(1) Performing time domain preprocessing on the obtained stereo signal to obtain a left channel signal obtained by the time domain preprocessing and a right channel signal obtained by the time domain preprocessing.

A stereo signal is collected by the collection component and sent to the encoding component 110 . Optionally, the collection component and encoding component 110 may be located on the same device. Alternatively, the collection component and encoding component 110 may be located on different devices.

The left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing are signals of two channels of the stereo signal obtained by preprocessing.

Optionally, pre-processing includes at least one of high-pass filtering, pre-emphasis, sampling rate conversion, and channel conversion. This is not limited in this embodiment.

(2) Perform delay estimation based on the left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing, and calculate the difference between the left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing. Get the inter-channel time difference.

(3) performing delay adjustment processing on the left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing based on the inter-channel time difference, and obtaining the left channel signal and delay matching processing obtained by delay matching processing; Obtain the right channel signal obtained by

(4) encoding the inter-channel time difference to obtain an encoding index of the inter-channel time difference;

(5) calculating the stereo parameters used in the time-domain downmixing process, encoding the stereo parameters used in the time-domain downmixing process, and obtaining the encoding index of the stereo parameters used in the time-domain downmixing process as obtain.

The stereo parameters used for time-domain downmix processing are used to perform time-domain downmix processing on the left channel signal obtained by delay matching processing and the right channel signal obtained by delay matching processing.

(6) performing time-domain downmix processing on the left channel signal obtained by delay matching processing and the right channel signal obtained by delay matching processing, based on the stereo parameters used in the time domain downmix processing, to obtain a primary Obtain a channel signal and a secondary channel signal.

A primary channel signal is used to represent information about the correlation between channels. Secondary channel signals are used for information about differences between channels. The secondary channel signal is minimal when the left channel signal obtained by the delay matching process and the right channel signal obtained by the delay matching process are matched in the time domain. In this case the stereo signal has the best effect.

(7) separately encode the primary channel signal and the secondary channel signal to obtain a first mono-encoded bitstream corresponding to the primary channel signal and a second mono-encoded bitstream corresponding to the secondary channel signal; .

(8) Write the encoding index of the inter-channel time difference, the encoding index of the stereo parameter, the first mono-encoded bitstream, and the second mono-encoded bitstream into the stereo-encoded bitstream.

Decoding component 120 is configured to decode the stereo-encoded bitstream produced by encoding component 110 to obtain a stereo signal.

Optionally, encoding component 110 is wired or wirelessly connected to decoding component 120, over which decoding component 120 obtains the stereo-encoded bitstream produced by encoding component 110. . Alternatively, encoding component 110 stores the generated stereo-encoded bitstream in memory, and decoding component 120 reads the stereo-encoded bitstream in memory.

Optionally, decoding component 120 may be implemented using software, or may be implemented using hardware, or may be implemented in the form of a combination of software and hardware. This is not limited in this embodiment.

Decoding component 120 decodes the stereo-encoded bitstream to obtain a stereo signal, which includes the following steps.

(1) Decoding the first mono-encoded bitstream and the second mono-encoded bitstream in the stereo-encoded bitstream to obtain a primary channel signal and a secondary channel signal.

(2) based on the stereo-encoded bitstream, obtain the coding indices of the stereo parameters used for the time-domain upmixing process, perform the time-domain upmixing process on the primary channel signal and the secondary channel signal, and obtain the time domain upmixing process; A left channel signal obtained by domain upmix processing and a right channel signal obtained by time domain upmix processing are obtained.

(3) obtaining a coding index for the inter-channel time difference based on the stereo coded bitstream, and performing delay adjustment on the left channel signal obtained by the time domain upmixing process and the right channel signal obtained by the time domain upmixing process; Go and get a stereo signal.

Optionally, encoding component 110 and decoding component 120 may be located on the same device or may be located on different devices. The device can be a mobile terminal with audio signal processing capabilities, such as a mobile phone, tablet computer, laptop portable computer, desktop computer, Bluetooth® speaker, pen recorder, or wearable device, or a core network or It may be a network element with audio signal processing capabilities in a wireless network. This is not limited in this embodiment.

For example, as shown in FIG. 2, in this embodiment encoding component 110 is located at mobile terminal 130, decoding component 120 is located at mobile terminal 140, and mobile terminal 130 and mobile terminal 140 are: Mutually independent devices with audio signal processing capabilities, such as mobile phones, wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, etc. Mobile terminals Description is made using an example where 130 is connected to mobile terminal 140 using a wireless or wired network.

Optionally, mobile terminal 130 includes collection component 131 , encoding component 110 and channel encoding component 132 . Collection component 131 is connected to encoding component 110 and encoding component 110 is connected to encoding component 132 .

Optionally, mobile terminal 140 includes audio reproduction component 141 , decoding component 120 and channel decoding component 142 . Audio playback component 141 is connected to decoding component 110 , which is connected to channel decoding component 132 .

After acquiring the stereo signal using the acquisition component 131, the mobile terminal 130 encodes the stereo signal using the encoding component 110 to obtain a stereo encoded bitstream, and then a channel encoding configuration. Encode the stereo-encoded bitstream using element 132 to obtain the transmitted signal.

Mobile terminal 130 transmits transmission signals to mobile terminal 140 using a wireless or wired network.

After receiving the transmitted signal, mobile terminal 140 decodes the transmitted signal using channel decoding component 142 to obtain a stereo-encoded bitstream, and uses decoding component 110 to obtain a stereo-encoded bitstream. Decode to obtain a stereo signal and reproduce the stereo signal using an audio reproduction component.

For example, as shown in FIG. 3, this embodiment provides an example in which encoding component 110 and decoding component 120 are located in network element 150 with audio signal processing capability within the same core network or wireless network. is used to explain.

Optionally, network element 150 includes channel decoding component 151 , decoding component 120 , encoding component 110 and channel encoding component 152 . Channel decoding component 151 is connected to decoding component 120 , decoding component 120 is connected to encoding component 110 , and encoding component 110 is connected to channel encoding component 152 .

After receiving a transmission transmitted by another device, channel decoding component 151 decodes the transmission to obtain a first stereo-encoded bitstream, and decoding component 120 decodes the stereo-encoded bitstream. to obtain a stereo signal, encoding component 110 encodes the stereo signal to obtain a second stereo coded bitstream, and channel coding component 152 encodes the second stereo coded bitstream to obtain a transmit signal. get

The other device may be a mobile terminal with audio signal processing capability or other network element with audio signal processing capability. This is not limited in this embodiment.

Optionally, encoding component 110 and decoding component 120 within the network element may transcode the stereo-encoded bitstream transmitted by the mobile terminal.

Optionally, in this embodiment, the device in which encoding component 110 is installed is referred to as an audio encoding device. In actual implementation, the audio encoding device may also have audio decoding functionality. This is not limited in this embodiment.

Optionally, this embodiment is described using only stereo signals as an example. In the present application, the audio encoding device may further process a multi-channel signal, the multi-channel signal comprising signals of at least two channels.

In order to facilitate understanding of the stereo signal encoding method in the embodiments of the present application, the following will first refer to FIG. 4 and FIG. The entire encoding process of the method is generally described.

FIG. 4 is a schematic flow diagram of a stereo signal encoding method in the frequency domain. This encoding method specifically includes 101-107.

101: Convert the time-domain stereo signal into a frequency-domain stereo signal.

102: Extract the frequency domain stereo parameters in the frequency domain.

103: Downmix processing is performed on the stereo signal in the frequency domain to obtain a downmix signal and a residual signal.

Downmix signals may also be referred to as central channel signals or primary channel signals, and parameter signals may be referred to as side channel signals or secondary channel signals.

104: Encoding the downmix signal to obtain encoding parameters corresponding to the downmix signal, and writing the encoding parameters into the encoded bitstream.

106: Encode the stereo parameters in the frequency domain to obtain encoding parameters corresponding to the stereo parameters in the frequency domain, and write the encoding parameters into the encoded bitstream.

In an optional implementation, the method may further include 105: encoding the residual signal to obtain encoding parameters corresponding to the residual signal, and writing the encoding parameters to the encoded bitstream.

107: Multiplex the bitstream.

FIG. 5 is a schematic flow chart of a stereo signal encoding method in the time-frequency domain. This encoding method specifically includes 201-208.

201: Perform time domain analysis on the stereo signal and extract time domain stereo parameters.

202: Transform the time-domain stereo signal into a frequency-domain stereo signal.

203: Extract the frequency domain stereo parameters in the frequency domain.

204: Perform downmix processing on the stereo signal in the frequency domain to obtain a downmix signal and a residual signal.

205: Encoding the downmix signal to obtain encoding parameters corresponding to the downmix signal, and writing the encoding parameters into the encoded bitstream.

207: Encoding the time-domain stereo parameters and the frequency-domain stereo parameters to obtain encoding parameters corresponding to the time-domain stereo parameters and encoding parameters corresponding to the frequency-domain stereo parameters, and encoding the encoding parameters Write to bitstream.

Optionally, the method further comprises 206: encoding the residual signal to obtain encoding parameters corresponding to the residual signal, and writing the encoding parameters to the encoded bitstream.

208: Multiplex the bitstream.

When the coding rate is relatively low, for example, when the coding bandwidth is Wideband, the coding rate is relatively low, Kilo-bytes per second (kbps), 16.4 kbps, 24. When it is 4kbps, or 32kbps, etc., the downmix signal of each frame of the stereo signal is coded to improve the sense of space and stability during playback of the stereo signal and to reduce the high frequency distortion of the stereo signal. Then, all residual signals in subbands satisfying a preset bandwidth range are coded. Alternatively, if the coding rate is relatively low, only the stereo parameters and the downmix signal are coded. Part or all of the residual signal is encoded only if the encoding rate is relatively high, such as 48 kbps, 64 kbps, or 96 kbps. The present application provides a stereo signal encoding method. In this way, the spatial perception and sound image stability of the decoded stereo signal can be improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the overall coding quality. be done.

FIG. 6 is a schematic flow diagram of a stereo signal encoding method 300 according to one embodiment of the present application. Method 300 may be performed by an encoder side, which may be an encoder or a device with stereo signal encoding capabilities. Method 300 includes the following steps.

The stereo signal encoding method of the present application may be a stereo encoding method that can be applied independently, or a stereo encoding method that is applied to multi-channel signal encoding. The encoder side processes the stereo signal frame by frame. In the following, using a wideband stereo signal where the signal length of each frame is 20ms as an example, and using the frame being processed by the encoder side (e.g. the current frame) as an example, the stereo signal encoding of the method 300 A detailed description will be given of the conversion method.

301: Determine a residual signal coding parameter of the current frame of the stereo signal based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame, and determine the residual signal of the current frame A signal coding parameter is used to indicate whether to code the residual signal of M subbands, where M subbands are at least part of the N subbands and N is greater than 1. A large positive integer, M≤N, where M is a positive integer.

Specifically, the encoder side divides the spectral coefficients of the current frame of the stereo signal to obtain N subbands, and divides the spectral coefficients of the current frame of the stereo signal into at least some of the N subbands (e.g., based on the downmix signal energy and the residual signal energy of each of the M subbands of M ≤ N), the encoder side determines the residual signal coding parameters of the current frame A signal coding parameter may be used to determine whether to code the residual signal for each of the M subbands.

302: Determine whether to code the residual signals of the M subbands of the current frame according to the residual signal coding parameters of the current frame.

Specifically, the encoder side encodes the residual signal of each of the M subbands of the current frame based on the residual signal coding parameters of the current frame determined in step 301. to decide whether

Upon determining to encode the residual signal of each of the M subbands, the downmix signal and residual signal of each of the M subbands are encoded.

If it is determined not to encode the residual signal for each of the M subbands, then the downmix signal for each of the M subbands is encoded.

In one implementation, by way of example and not limitation, the M subbands are the M subbands within the N subbands whose subband index numbers are less than a preset maximum subband index number. In other words, the M subbands are subbands with relatively low frequencies within the N subbands, specifically, the frequencies of the M subbands are equal to M lower than the frequencies of the NM subbands other than the subbands.

Specifically, since different maximum subband index numbers are preconfigured based on different coding rates, M subbands whose subband index numbers are less than or equal to the preconfigured maximum subband index number are preconfigured. One of the N subbands is selected based on the configured maximum subband index number, and the residual signal coding parameters of the current frame are determined based on the M subbands.

For example, if the coding rate is 26 kbps, N=10, M=5, and the preset maximum subband index number is set to 4, this means that the residual signal coding parameters of the current frame are It is determined based on 5 subbands with subband index numbers 0 to 4 in 10 subbands.

In another example, if the coding rate is 44 kbps, N=12, M=6, and the preset maximum subband index number is set to 5, this is the residual signal coding for the current frame. It shows that the parameters are determined based on 6 subbands with subband index numbers 0 to 5 in 12 subbands.

In another example, if the coding rate is 56 kbps, N=12, M=7, and the preset maximum subband index number is set to 6, this is the residual signal coding for the current frame. It shows that the parameters are determined based on 7 subbands with subband index numbers 0 to 6 in 12 subbands.

In other implementations, for different coding rates, the maximum subband index number and the minimum subband index number of the M subbands at different coding rates can be preset, so the subband index numbers are preset N subbands that are greater than or equal to the minimum subband index number and less than or equal to the preset maximum subband index number are N based on the preset maximum subband index number and the preset minimum subband index number , and the residual signal coding parameters for the current frame are determined based on the M subbands.

For example, if the coding rate is 26 kbps, N=10, M=4, the preset minimum subband index number is set to 4, and the preset maximum subband index number is set to 7, This indicates that the residual signal coding parameters of the current frame are determined based on four subbands with subband index numbers 4 to 7 among the ten subbands.

By way of example and not limitation, determining whether to encode the residual signal of each of the M subbands based on the residual signal coding parameters of the current frame comprises: determining whether to encode the residual signal of based on the result of comparing the residual signal coding parameter of the current frame with a preset first threshold, wherein the first threshold is A step greater than 0 and less than 1.0, and determining not to encode the residual signal of each of the M subbands if the residual signal coding parameter of the current frame is less than or equal to the first threshold. or determining to encode the residual signal of each of the M subbands if the residual signal coding parameter is greater than a first threshold.

Specifically, the encoder side compares the residual signal coding parameter of the current frame with a preset first threshold, and the residual signal coding parameter of the current frame is greater than the first threshold. If so, determine to encode the residual signal of each of the M subbands, or if the residual signal encoding parameter of the current frame is less than or equal to the first threshold, determine to encode the residual signal of each of the M subbands. Decide not to encode the residual signal.

For example, in one implementation, the first threshold is 0.075. If the value of the residual signal coding parameter of the current frame is 0.06, the encoder side does not code the residual signal of each of the M subbands.

It should be appreciated that the value of the first threshold is only an example and that the first threshold may alternatively be other values greater than 0 and less than 1.0. For example, the first threshold is 0.55, 0.46, 0.86, or 0.9.

In another optional implementation, the encoder side may use 0 or 1 to further indicate the comparison result between the residual signal coding parameter of the current frame and the first threshold. For example, 0 is used to indicate that the residual signal of each of the M subbands should not be coded, and 1 is used to indicate that the residual signal of each of the M subbands is coded. Used to indicate that something should be done. Of course, 1 may alternatively be used to indicate that the residual signal in each of the M subbands should not be encoded, and 0 alternatively may be used to indicate that each residual signal of is to be encoded.

In the following, in order to explain in detail how the encoder side determines the residual signal coding parameters of the current frame, M subbands are assigned the maximum subband index number ( For example, the maximum subband index number is M−1).

Method 1

The encoder side determines the residual signal coding parameters of the current frame based on the downmix signal energy, residual signal energy and side gain of each of the M subbands.

In one possible implementation, the encoder side determines the first parameter based on the downmix signal energy, the residual signal energy, and the side gain for each of the M subbands, the first parameter being , denoting the value relationship between the downmix signal energy and the residual signal energy for each of the M subbands,
determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the second parameter being a value relationship between the first energy sum and the second energy sum where the first energy sum is the sum of the residual signal energies and the downmix signal energies of the M subbands, and the second energy sum is the M is the sum of the residual signal energy and the downmix signal energy of the subbands of , where the M subbands of the current frame have the same subband index numbers as the M subbands of the previous frame, and
A residual signal coding parameter of the current frame is finally determined based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame.

Specifically, when determining the first parameter, the encoder side determines M energy parameters based on the downmix signal energy, residual signal energy, and side gain of each of the M subbands. and M energy parameters each indicating a value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters representing the M subbands. In one-to-one correspondence with the band, the encoder side finally determines the energy parameter having the maximum value among the M energy parameters as the first parameter.

Optionally, the energy parameter of the subband with subband index number b among the M energy parameters may be determined using the function:
res_dmx_ratio[b] = f(g(b), res_cod_NRG_M[b], res_cod_NRG_S[b]) (1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. where g(b) represents the function of the side gain side_gain[b] of the subband with subband index number b.

Specifically, in one embodiment, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies the following equation.
res_dmx_ratio[b] = res_cod_NRG_S[b] / (res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b] + 1) (2)

The first parameter is denoted res_dmx_ratio, where res_dmx_ratio satisfies the following equation.
res_dmx_ratio = max(res_dmx_ratio[0], res_dmx_ratio[1], ..., res_dmx_ratio[M-1]) (3)

When determining the second parameter, the encoder side first separately determines the sum of the residual signals of the M subbands and the sum of the downmix signals of the M subbands; dmx_nrg_all_curr and the sum of residual signals of M subbands is denoted as res_nrg_all_curr.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the formula:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

It should be appreciated that the value of γ ₁ is only an example and that the value of γ ₁ may alternatively be other values between 0 and 1 inclusive. For example, γ ₁ is 0.3, 0.5, 0.6, or 0.8.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the formula:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

It should be appreciated that the value of γ ₂ is only an example and that the value of γ ₂ may alternatively be other values between 0 and 1 inclusive. For example, _γ2 is 0.2, 0.5, 0.7, or 0.9.

The encoder side determines the sum of the downmix signal energy and the residual signal energy (ie, the first energy sum) of the M subbands of the current frame based on dmx_nrg_all_curr and res_nrg_all_curr. The first energy sum is denoted dmx_res_all.

Optionally, dmx_res_all satisfies the following formula.
dmx_res_all = res_nrg_all_curr + dmx_nrg_all_curr(6)

The encoder side may further determine the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency domain signal of the frame previous to the current frame (i.e., the second energy sum). Well, the M subbands of the frame before the current frame have the same subband index number as the M subbands. The second energy sum is denoted dmx_res_all_prev.

For the determination of the second energy sum dmx_res_all_prev, please refer to the method for determining the first energy sum dmx_res_all described above. For the sake of brevity, the details are not repeated here.

After determining the first energy sum and the second energy sum, the encoder side may determine a second parameter based on the first energy sum and the second energy sum.

Optionally, the second parameter is the frame-to-frame energy variation rate, and the frame-to-frame energy variation rate is denoted as frame_nrg_ratio.

Optionally, in one implementation, the frame-to-frame energy variation rate frame_nrg_ratio satisfies the following equation.
frame_nrg_ratio = dmx_res_all/dmx_res_all_prev(7)

Optionally, in other implementations, the frame-to-frame energy variation rate frame_nrg_ratio satisfies the following equation.
frame_nrg_ratio = min(5.0, max(0.2, dmx_res_all/dmx_res_all_prev)) (8)

The max function is used to return the larger value at the given parameters (0.2, frame_nrg_ratio_prev) and the min function is used to return the larger value at the given parameters (5.0, max(0.2, frame_nrg_ratio_prev)) Used to return the minimum value. Compared to Eq. (7), Eq. (8) has an additional correction operation, so the frame_nrg_ratio determined using Eq. (8) represents the inter-frame energy variation between the current and previous frames. can be better reflected.

After determining the first parameter and the second parameter, the encoder side calculates the residual of the current frame based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame. Difference signal encoding parameters may be determined.

By way of example and not limitation, the residual signal coding parameters for the current frame may be long-term smoothing parameters for the current frame. In other words, the encoder side determines the long-term smoothing parameter of the current frame based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame, and then M A long-term smoothing parameter of the current frame may be compared to a preset first threshold to determine whether to code the residual signal of each of the subbands of .

For example, the long-term smoothing parameter for the current frame satisfies
res_dmx_ratio_lt = res_dmx_ratio α + res_dmx_ratio_lt_prev (1 - α) (9)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 is.

When res_dmx_ratio_lt is calculated according to equation (9), if the value of the first parameter and/or the value of the second parameter changes, the value of parameter α in equation (9) may change accordingly. In other words, when the value of the first parameter and/or the value of the second parameter change, the weight of the long-term smoothing parameter of the frame before the current frame in equation (9) may change accordingly. .

For example, if the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is equal to the preset value of the first parameter. greater than the value of α for greater than or equal to the second threshold, where the second threshold is between 0 and 0.6, and the third threshold is between 2.7 and 3.7, or the second parameter is less than a preset fifth threshold, the value of α if the first parameter is greater than a preset fourth threshold is less than or equal to the preset fourth threshold the fourth threshold is between 0 and 0.9 inclusive, the fifth threshold is between 0 and 0.71 inclusive, or the second threshold with the first parameter preset If the value of α is less than the threshold and the second parameter is greater than the preset third threshold, the second parameter is greater than or equal to the preset fifth threshold and the preset third is greater than the value of α when it is less than or equal to the threshold, the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0 .71 or less.

For example, the value of the second threshold may be 0.1 and the value of the third threshold may be 3.2. Specifically, if the second parameter frame_nrg_ratio is greater than 3.2, the The value of α when the parameter res_dmx_ratio of 1 is less than 0.1 is greater than the value of α when res_dmx_ratio is greater than or equal to 0.1, or the value of the fourth threshold is 0.4 and the value of the fifth threshold may be 0.21. Specifically, when frame_nrg_ratio is less than 0.21, the value of α when res_dmx_ratio is greater than 0.4 is the value of α when res_dmx_ratio is 0.4 or less. or the second threshold value is 0.1, the third threshold value is 3.2, and the fifth threshold value is 0.21, specifically the value of α when res_dmx_ratio is less than 0.1 and frame_nrg_ratio is greater than 3.2 is greater than the value of α when frame_nrg_ratio is between 0.21 and 3.2, or the fourth threshold may be 0.4 and the value of the fifth threshold may be 0.21, specifically, the value of α when res_dmx_ratio is greater than 0.4 and frame_nrg_ratio is less than 0.21 is greater than the value of α when frame_nrg_ratio is 0.21 or more and 3.2 or less.

Further, for example, when res_dmx_ratio is less than 0.1 and frame_nrg_ratio is greater than 3.2, the value of α is 0.5, or when frame_nrg_ratio is between 0.21 and 3.2, the value of α is 0.1.

Note that the stated second through fifth threshold values and α values are illustrative examples only and do not constitute any limitations on the present application. The second through fifth threshold values and the value of α may alternatively be other values in a given interval.

Note further that the current frame has no previous frame if it is the first frame processed by the encoder side. In this case, when the long-term smoothing parameter of the current frame is determined, the long-term smoothing parameter of the frame before the current frame in the above formula is the preset long-term smoothing parameter. By way of example and not limitation, the value of the preset long-term smoothing parameter may be 1.0, or of course other values such as 0.9 or 1.1.

Method 2

The method for determining the residual signal coding parameter in Method 2 is similar to the method in Method 1, the difference being that the method for determining the first parameter is different. Reference may be made to the related discussion of determining signal coding parameters. For the sake of brevity, only the method for determining the first parameter in Method 2 is described herein.

By way of example and not limitation, the encoder side determines a first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being the 2 shows the value relationship between the downmix signal energy and the residual signal energy for each of .

Specifically, when determining the first parameter, the encoder side determines M energy parameters based on the downmix signal energy and the residual signal energy of each of the M subbands; energy parameters each indicate a value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters are paired with the M subbands. 1, and the encoder side finally determines the energy parameter having the maximum value among the M energy parameters as the first parameter.

Optionally, the energy parameter of the subband whose subband index number is b among the M energy parameters determined by the encoder side may be determined using the following function:
res_dmx_ratio[b] = f(res_cod_NRG_M[b], res_cod_NRG_S[b]) (10)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. show.

For example, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies the following equation.
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b] (11)

The first parameter is denoted res_dmx_ratio, where res_dmx_ratio satisfies the following equation.
res_dmx_ratio = max(res_dmx_ratio[0], res_dmx_ratio[1], ..., res_dmx_ratio[M-1]) (12)

After determining the first parameter, the encoder side determines the second parameter according to the method described in Method 1, finally determines the residual signal coding parameter according to the method described in Method 1, and Based on the difference signal coding parameters, it may be determined whether to code the residual signal for each of the M subbands.

Method 3

The method for determining the residual signal coding parameter in method 3 is similar to the method in method 1, the difference being that the method for determining the first parameter is different. Reference may be made to the related discussion of determining signal coding parameters. For the sake of brevity, only the method for determining the first parameter in Method 3 is described herein.

By way of example and not limitation, the encoder side determines a first parameter based on downmix signal energy and residual signal energy for each of the M subbands, corrects the first parameter, and final As the first parameter, determine the first parameter obtained by the correction, the first parameter indicating the value relationship between the downmix signal energy and the residual signal energy for each of the M subbands .

Specifically, when determining the first parameter, the encoder side determines M energy parameters based on the downmix signal energy and the residual signal energy of each of the M subbands; energy parameters each indicate a value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters are paired with the M subbands. 1, and the encoder side determines the sum of M energy parameters as the first parameter.

Optionally, the energy parameter of the subband whose subband index number is b among the M energy parameters determined by the encoder side may be determined using function (1).

For example, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies Equation (2).

Optionally, the energy parameter of the subband whose subband index number is b among the M energy parameters determined by the encoder side may be determined using function (11).

For example, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies Equation (11).

For example, the first parameter res_dmx_ratio ₁ determined by the encoder side based on M energy parameters satisfies the following equation.

In addition, the encoder side may further determine the maximum value res_dmx_ratio_max among the M energy parameters, where res_dmx_ratio_max satisfies Equation (12).

The encoder side corrects res_dmx_ratio ₁ based on res_dmx_ratio_max of each of the M subbands and downmix signal energy res_cod_NRG_M[b], and determines res_dmx_ratio ₂ obtained by the correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2) (15)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

After determining the first parameter obtained by correction (res_dmx_ratio ₃ obtained by correction), the encoder side determines the second parameter according to the method described in Method 1, and determines the second parameter according to the method described in Method 1. A residual signal coding parameter may be finally determined, and whether to code the residual signal for each of the M subbands may be determined based on the residual signal coding parameter.

Method 4

The method for determining the residual signal coding parameter in method 4 is similar to the method in method 1, the difference being that the method for determining the first parameter is different. Reference may be made to the related discussion of determining signal coding parameters. For the sake of brevity, only the method for determining the first parameter in method 4 is described herein.

By way of example and not limitation, the encoder side determines the first parameter based on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands.

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

Optionally, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies equation (4). index number

Optionally, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies equation (5). index number

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr(16)

After determining the first parameter, the encoder side determines the second parameter according to the method described in Method 1, finally determines the residual signal coding parameter according to the method described in Method 1, and Based on the difference signal coding parameters, it may be determined whether to code the residual signal for each of the M subbands.

In order to better understand the whole encoding of stereo signals, in the following, we will use a wideband stereo signal as an example where the signal length of each frame is 20ms, and the frame being processed by the encoder side (e.g. the current frame ) as an example, the stereo signal encoding method 300 of this embodiment of the present application will be described with reference to FIG. The stereo signal encoding method shown in FIG. 7 includes at least the following steps.

401: Perform time domain preprocessing on the left channel time domain signal and the right channel time domain signal to obtain a left channel time domain signal and a right channel time domain signal obtained by the time domain preprocessing.

Specifically, the signal length of the current frame is 20 ms. If the sampling frequency is 16 kHz (KHz), after sampling, the frame length of the current frame H=320, in other words the current frame contains 320 sampling points.

The current frame stereo signal includes a current frame left channel time domain signal and a current frame right channel time domain signal. The left channel time domain signal of the current frame is denoted as x _L (n) and the right channel time domain signal of the current frame is denoted as x _R (n). n is the sequence number of sampling points, n=0, 1, . . . , and H−1. Left channel time domain signals and right channel time domain signals may be referred to as left and right channel time domain signals.

The step of performing time-domain preprocessing on the left-channel time-domain signal and the right-channel time-domain signal of the current frame includes converting the left-channel time-domain signal and the right-channel time-domain signal of the current frame obtained by the time-domain preprocessing into: performing high-pass filtering on the left and right channel time-domain signals of the current frame, respectively, to obtain. The left channel time domain signal of the current frame obtained by preprocessing is denoted as x _{L_HP} (n), and the right channel time domain signal of the current frame obtained by preprocessing is denoted as x _{R_HP} (n). n is the sequence number of sampling points, n=0, 1, . . . , and H−1. The left channel time domain signal of the current frame obtained by time domain preprocessing and the right channel time domain signal of the current frame obtained by time domain preprocessing are the left and right channels of the current frame obtained by time domain preprocessing. It can be called a time domain signal. An Infinite Impulse Response (IIR) digital filter with a cutoff frequency of 20 Hz (Hz) may be used during the high-pass filtering process, or other types of filters may be used.

For example, when the sampling rate of a stereo signal is 16kHz, the corresponding transfer function of a high-pass filter with a cutoff frequency of 20Hz can be:

b0 = 0.994461788958195, b1 = -1.988923577916390, b2 = _{0.994461788958195} , a1 _{= 1.98892905899653} , _{a2 =} -0.988954249933127, and z represents the transform coefficient of the Z transform. The corresponding time domain filters are:
x _{L_HP} (n) = b ₀ x _L (n) + b ₁ x _L (n-1) + b ₂ x _L (n-2) - a ₁ x _{L_HP} (n-1) - a ₂ x _{L_HP} (n-2) (18)

402: Perform time domain analysis on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing.

Specifically, time domain analysis may include transient detection and the like. Transient detection separately performs energy detection on the left channel time domain signal and right channel time domain signal of the current frame obtained by preprocessing to detect whether an energy burst occurs in the current frame. It can be

For example, the energy E _{cur_L} of the left channel time domain signal of the current frame obtained by preprocessing is calculated. Transient detection uses the energy E _{pre_L} of the left channel time domain signal of the frame before the current frame obtained by preprocessing to obtain the transient detection result of the left channel time domain signal of the current frame obtained by preprocessing. It is based on the absolute value of the difference between the energy _{Ecur_L} of the left channel time domain signal of the current frame obtained by preprocessing. Transient detection can be performed on the right channel time domain signal of the current frame obtained by preprocessing using the same method.

The time domain analysis may include other prior art time domain analyzes in addition to transient detection. For example, time domain analysis may include time domain Inter-channel Time Difference (ITD) parameter determination, time domain delay matching processing, and band extension preprocessing.

403: Perform a time-frequency transform on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing to obtain a left channel frequency domain signal and a right channel frequency domain signal.

Specifically, a discrete Fourier transform may be performed on the left channel time domain signal obtained by the time domain preprocessing to obtain the left channel frequency domain signal, and to obtain the right channel frequency domain signal: A discrete Fourier transform is performed on the right channel time domain signal obtained by the time domain preprocessing.

To overcome the problem of spectral aliasing, a convolution-add method may be used to process the Discrete Fourier Transform between two consecutive times, in some cases adding zeros to the input signal of the Discrete Fourier Transform. obtain.

The discrete Fourier transform may be performed once per frame, or each frame of the signal may be divided into P subframes, where P is a positive integer greater than or equal to 2, and the discrete Fourier transform may be performed once per subframe. is performed once every

For example, the Discrete Fourier Transform is performed once for the current frame, the left channel frequency domain signal of the current frame on which the Discrete Fourier Transform is performed is denoted as L(k), and the current frame on which the Discrete Fourier Transform is performed is denoted R(k). k represents the frequency bin index number, k=0, 1, . The current frame being read contains L frequency bins.

In another example, the current frame of the signal is partitioned into P subframes, where P is a positive integer greater than or equal to 2. The left channel frequency domain signal of the subframe where the discrete Fourier transform is performed with index number _i is denoted by Li(k), and the right channel frequency of the subframe where the discrete Fourier transform is performed with index number i The area signal is denoted R _i (k). i represents the subframe index number, i = 0, 1, ..., P-1, k represents the frequency bin index number, k = 0, 1, ..., L-1, and L is , represents the frame length of each subframe on which the discrete Fourier transform is performed, in other words, each subframe on which the discrete Fourier transform is performed contains L frequency bins.

404: Determine ITD parameters and encode the determined ITD parameters.

Specifically, there are multiple methods for determining the ITD parameters. The ITD parameters may be determined only in the frequency domain, only in the time domain, or determined in the time-frequency domain. This is not a limitation in this application.

および

が計算される。 ITD parameters can be extracted in the time domain using cross-correlation coefficients. For example, in the range _0≤i≤Tmax ,

and

is calculated.

の場合、ITDパラメータ値は、max（c_n（i））に対応するインデックス番号の反対の数である。

の場合、ITDパラメータ値は、max（c_p（i））に対応するインデックス番号である。iは、相互相関係数を計算するためのインデックス番号を表し、jは、サンプリング点のインデックス番号を表し、T_maxは、異なるサンプリングレートにおけるITD値の最大値に対応し、Hは、現在のフレームのフレーム長を表す。

, the ITD parameter value is the opposite number of the index number corresponding to max(c _n (i)).

, the ITD parameter value is the index number corresponding to max(c _p (i)). i represents the index number for calculating the cross-correlation coefficient, j represents the index number of the sampling point, T _max corresponds to the maximum ITD value at different sampling rates, H is the current Represents the frame length of the frame.

The ITD parameters may alternatively be determined in the frequency domain based on the left channel frequency domain signal and the right channel frequency domain signal. For example, time-domain A signal may be transformed into a frequency domain signal.

In this embodiment of the present application, the left channel frequency domain signal of the subframe whose index number is i and the discrete Fourier transform is performed is denoted as L _i (k), where k=0, 1, . . . , L/ 2−1, the index number is i, and the right channel frequency domain signal of the subframe under transformation is denoted R _i (k), where k=0, 1, . . . , L/2−1 and i=0, 1, . . . , P−1. The frequency-domain correlation coefficient of the subframe with index number i is calculated according to XCORR _i (k) = L _i (k) R ^* _i (k), where R ^* _i (k) is transformed Represents the conjugate of the right channel frequency domain signal for the ith subframe.

であることを得るために、L／2－T_max≦n≦L／2＋T_maxの範囲でxcorr_i（n）の最大値が探索される。 The frequency domain cross-correlation coefficients are transformed to the time domain xcorr _i (n), where n=0, 1, .

The maximum value of xcorr _i (n) is searched for in the range L/2-T _max ≤ n ≤ L/2 + T _max to obtain that .

に従って振幅値がさらに計算されてもよく、ITDパラメータ値は

であり、具体的には、ITDパラメータ値は、最大振幅値に対応するインデックス番号である。 In addition, in the search range −T _max ≤ j ≤ T _max , based on the left channel frequency domain signal and the right channel frequency domain signal of the subframe where the index number is i and the DFT transformation is performed

Amplitude values may be further calculated according to and the ITD parameter value is

and specifically, the ITD parameter value is the index number corresponding to the maximum amplitude value.

After the ITD parameters are determined, the ITD parameters may be encoded to obtain the encoding parameters, which are written into the stereo encoded bitstream.

405: Perform time shift adjustment on the left frequency domain signal and the right channel frequency domain signal based on the ITD parameters.

Specifically, time shift adjustments may be made to the left channel frequency domain signal and the right channel frequency domain signal according to any technique. This is not a limitation in this embodiment of the application.

For example, the current frame of the signal is divided into P subframes, where P is a positive integer greater than or equal to 2. The left channel frequency domain signal obtained by time shift adjustment of the subframe with index number i may be denoted as L' _i (k), where k = 0, 1, ..., L/2-1. , and the right channel frequency domain signal obtained by time shift adjustment of the subframe with index number i may be denoted as R′ _i (k), where k represents the frequency bin index number and k= 0, 1, . . . , L/2−1, i represents the subframe index number, and i=0, 1, .

T _i represents the ITD parameter value of the subframe with index number i, L represents the length of the subframe over which the discrete Fourier transform is performed, L _i (k) is the index number i, R i (k) represents the left channel frequency domain signal of the ith subframe in which the transform is performed, R _i (k) represents the right channel frequency domain signal of the subframe in which the index number is i, where i is Represents the subframe index number, i=0, 1, . . . , P−1.

406: Calculate another frequency domain stereo parameter based on the left channel frequency domain signal and the right channel frequency domain signal obtained by the time shift adjustment, and encode the other frequency domain stereo parameter.

Specifically, the other frequency-domain stereo parameters are Inter-channel Phase Difference (IPD) parameters, and/or Inter-channel Level Difference (ILD) parameters, and/or It may include, but is not limited to, sub-band side gains and the like. ILD may also be referred to as inter-channel amplitude difference.

After obtaining the other frequency-domain stereo parameters by calculation, the other frequency-domain stereo parameters may be coded to obtain the coding parameters, and the coding parameters are written into the stereo-coded bitstream.

407: Determine M subbands that satisfy a preset condition from the N subbands included in the frequency domain signal of the current frame.

Specifically, the frequency domain signal obtained by the time shift adjustment of the current frame is divided into subbands. For example, the frequency domain signal of the current frame is divided into N subbands (where N is a positive integer greater than or equal to 2), and the frequency bin contained in the subband with subband index number b is k ∈ [ band_limits(b), band_limits(b+1)−1], where band_limits(b) represents the minimum index number of frequency bins included in the subband whose subband index number is b, and band_limits(b+1) represents the subband It represents the minimum index number of the frequency bins included in the subband whose band index number is b+1. According to a preset condition, M subbands that satisfy the preset condition are determined among the N subbands.

For example, the preset condition is that the subband index number is less than or equal to the preset maximum subband index number, i.e., b≤res_cod_band_max, where res_cod_band_max represents the preset maximum subband index number; can be

Alternatively, the preset condition is that the subband index number is less than or equal to the maximum preset subband index number and greater than or equal to the minimum preset subband index number, i.e. res_cod_band_min≤b≤res_cod_band_max, res_cod_band_max is , res_cod_band_min represents the preset maximum subband index number, and res_cod_band_min represents the preset minimum subband index number.

Furthermore, for wideband stereo signals, different preset conditions may be set based on different coding rates. For example, when the coding rate is 26 kbps, the preset condition is subband index number b≦5, in other words, the preset maximum subband index number is 5. When the coding rate is 44 kbps, the preset condition is subband index number b≦6, in other words, the preset maximum subband index number is 6. When the coding rate is 56 kbps, the preset condition is subband index number b≦7, in other words, the preset maximum subband index number is 7.

For example, if the preset condition is subband index number b≤4, the 5 subbands with index numbers 0 to 4 are the subbands satisfying the preset condition among the N subbands of the current frame. can be determined as a band.

In addition, if the current frame of the signal is divided into P subframes (P is a positive integer greater than or equal to 2), each subframe obtained by time shift adjustment is divided into subbands. For example, a subframe with an index number of i (i = 0, 1, ..., P-1) is divided into N subbands, and subbands with an index number of b in the subframe with an index number of i The frequency bins included in the band are k _i ∈ [band_limits(b), band_limits(b+1)−1], where band_limits(b) is the subband with index number b in the subframe with index number i. and band_limits(b+1) represents the minimum index number of frequency bins included in the subband with index number b+1 in the subframe with index number i.

According to a preset condition, M subbands that satisfy the preset condition are determined from among the N subbands included in each frame.

The preset condition may be that the subband index number is greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number, ie, res_cod_band_min≤b≤res_cod_band_max.

For example, if the preset condition is 4≤b≤8, the 5 subbands with index numbers 4 to 8 satisfy the preset condition among the N subbands in each subframe. is determined as

408: Based on the left channel frequency domain signal and the right channel frequency domain signal obtained by the time shift adjustment, calculate the sub-band downmix signal and the residual signal that satisfy the preset condition.

Specifically, the method for calculating the subband downmix signal and residual signal satisfying the preset condition is that the current frame has P subframes (P is a positive integer greater than or equal to 2). It will be described using an example of partitioning (eg, the current frame may be partitioned into 2 subframes or 4 subframes).

For example, if the preset condition is that the subband index number b is less than or equal to 5, the downmix signal and residual signal of the subbands with index numbers from 0 to 5 in each subframe are calculated. .

RES_i’（k）＝RES_i（k）－g＿ILD_i・DMX_i（k）（21）

β＝arctan（sin（IPD_i（b）），cos（IPD_i（b））＋2・c）（24）、および

A downmix signal of a subband whose index number is b (b≤5) in the subframe whose index number is i is denoted as DMX _i (k), and the index number in the subframe whose index number is i is The residual signal of subband b is denoted RES _i '(k), where DMX _i (k) and RES _i '(k) satisfy the following equations.

RES _i '(k) = RES _i (k) - g_ILD _i DMX _i (k) (21)

β = arctan(sin(IPD _i (b)), cos (IPD _i (b)) + 2 c) (24), and

IPD _i (b) represents the IPD parameters of the subband whose index number is b in the subframe whose index number is i, and g_ILD _i is the index number of b in the subframe whose index number is i. represents the side gain of a subband, L' _i (k) is the left channel frequency domain signal obtained by time shift adjustment of the subband with index number b in the subframe with index number i , R′ _i (k) represents the right channel frequency domain signal obtained by time shift adjustment of the subband whose index number b is in the subframe whose index number is i, and L″ _i (k) represents the left channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number i, and R'' _i ( k) represents the right channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number i, where i is the subframe index number; , where i=0, 1, . , represents the minimum index number of frequency bins included in the subband with index number b in the subframe with index number i, and band_limits(b+1) represents the minimum index number of the frequency bin in the subframe with index number i. Represents the lowest index number of frequency bins contained in the subband that is b+1.

In another example, the downmix signal DMX _i (k) for the subband with index number b in the subframe with index number i may alternatively be calculated according to the following method.
DMX _i (k) = [L''(k) + R''(k)] c(26), and

L'' _i (k) represents the left channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number i, and R'' _i (k) represents the right channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number i, i is the sub represents the frame index number, i=0, 1, . b) represents the minimum index number of frequency bins included in the subband whose subband index number is b, and band_limits(b+1) is the subband whose index number is b+1 in the subframe whose index number is i. represents the lowest index number of the frequency bins contained in . The method for calculating downmix signal energy and residual signal energy is not limited in this embodiment of the application.

409: Determining residual signal coding parameters based on the downmix signal energy and the residual signal energy of the sub-bands satisfying a preset condition.

410: Determine whether the residual signal of each of the M subbands of the current frame needs to be coded based on the residual signal coding parameters. If it is determined that the residual signal needs to be encoded, 412 is performed. If it is determined that the residual signal does not need to be encoded, 411 is performed.

411: Encoding the downmix signal of each of the M subbands of the current frame based on the residual signal encoding parameters. In this case the residual signal does not need to be coded.

412: Encode the downmix signal and the residual signal for each of the M subbands of the current frame based on the residual signal coding parameters.

See the related description of method 300 for specific implementations of steps 409 through 411 . For the sake of brevity, the details are not repeated here.

In method 400, the encoder side divides the current frame into P subframes, where P is a positive integer greater than or equal to 2, and divides the spectral coefficients of each of the P subframes into N subbands. , and a downmix signal of M subbands (the M subbands are part of at least the N subbands) in each subframe, where the residual signal coding parameter satisfies a preset condition energy and residual signal energy, then in method 300 the residual signal energy res_cod_NRG_S[b] for the subband with index number b in the current frame is is the sum of the residual signal energy of the subband with index number b in the current frame, and the downmix signal energy res_cod_NRG_M[b] of the subband with index number b in the current frame is the sum of all P subframes is the sum of the downmix signal energies of the subband whose index number is b.

For example, the current frame is divided into two subframes and the spectral coefficients of each of the two subframes are divided into N subbands. Therefore, in method 300, the downmix signal energy res_cod_NRG_M[b] of the subband with index number b in the current frame is equal to the downmix signal energy of the subband with index number b in subframe 1. The sum of the downmix signal energy of the subband with index number b in frame 2, and the residual signal energy res_cod_NRG_S[b] of the subband with index number b in the current frame is is the sum of the residual signal energy of the subband with index number b in subframe 2 and the residual signal energy of the subband with index number b in subframe 2 .

The stereo signal encoding method according to the embodiment of the present application has been described in detail above with reference to FIGS. 1 to 7. FIG. The stereo signal encoding device in the embodiment of the present application will be described below with reference to FIGS. 8 and 9. FIG. It should be understood that both the devices in FIGS. 8 and 9 correspond to the stereo signal encoding method in the embodiments of the present application. In addition, any of the devices in Figures 8 and 9 can perform the stereo signal encoding method in the embodiments of the present application. For the sake of brevity, repetitive descriptions are omitted where appropriate.

FIG. 8 is a schematic block diagram of a stereo signal encoding device according to one embodiment of the present application. Apparatus 500 of FIG.
A first determination configured to determine a residual signal coding parameter for the current frame of the stereo signal based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame. A module 501, wherein the residual signal coding parameter of the current frame is used to indicate whether to code the residual signal of M subbands, where M subbands are N subbands. a first determining module 501 that is at least part of a band, where N is a positive integer greater than 1, M≦N, where M is a positive integer;
a second decision module 502, configured to decide whether to code the residual signal of the M subbands of the current frame based on the residual signal coding parameters of the current frame; include.

In this application, the residual signal coding parameters are determined based on the downmix signal energy and the residual signal energy of M subbands within the N subbands that satisfy a preset bandwidth range. , whether to code the residual signal of each of the M subbands is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Optionally, in one implementation, the M subbands are the M subbands whose subband index numbers are less than or equal to a preset maximum subband index number in the N subbands.

Optionally, in one embodiment, the M subbands have a subband index number greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number in the N subbands. are M subbands.

Optionally, in one implementation, the second determining module 502 compares the residual signal coding parameter to a preset first threshold, if the first threshold is greater than 0 and less than 1.0, determining not to encode the residual signal for each of the M subbands if the residual signal coding parameter is less than or equal to the first threshold, or the residual signal coding parameter is greater than the first threshold; If so, determine to encode the residual signal of each of the M subbands.

Optionally, in one implementation, the first determining module 501 determines the residual signal coding parameters based on the downmix signal energy, the residual signal energy and the side gains for each of the M subbands. is further configured as

Optionally, in one implementation, the first determining module 501 determines the first parameter based on the downmix signal energy, the residual signal energy and the side gain for each of the M subbands; A parameter of 1 indicates a value relationship between the downmix signal energy and the residual signal energy of each of the M subbands, based on the downmix signal energy and the residual signal energy of each of the M subbands. to determine a second parameter, the second parameter indicating a value relationship between the first energy sum and the second energy sum, the first energy sum being the residual signal energy of the M subbands and the downmix signal energy, the second energy sum is the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency domain signal of the previous frame of the current frame, and the current where the M subbands of the frame have the same subband index numbers as the M subbands of the previous frame, and the long-term smoothing of the first parameter, the second parameter, and the frame before the current frame It is further configured to determine residual signal coding parameters based on the parameters.

Optionally, in one implementation, the first determining module 501 determines M energy parameters based on downmix signal energy, residual signal energy, and side gain for each of the M subbands; The M energy parameters each indicate a value relationship between the downmix signal energy and the residual signal energy for one of the M subbands, and the M energy parameters correspond to the M subbands. It is further configured to determine the energy parameter having the largest value among the M energy parameters with a one-to-one correspondence as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by the first determination module 501 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/(res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g(b) is 4 represents a function of the side gain side_gain[b] of the subband whose subband index number is b.

Optionally, in one implementation, the first determining module 501 determines the first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being , a value relationship between the downmix signal energy and the residual signal energy for each of the M subbands, and based on the downmix signal energy and the residual signal energy for each of the M subbands, a second determining a parameter, the second parameter indicating a value relationship between the first energy sum and the second energy sum, the first energy sum being the residual signal energy of the M subbands and the downmix signal; is the sum of the energies of the current frame, and the second energy sum is the sum of the residual signal energies and the downmix signal energies of the M subbands in the frequency-domain signal of the previous frame of the current frame, and the M subbands have the same subband index numbers as the M subbands of the previous frame, based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame It is further configured to determine residual signal coding parameters.

Optionally, in one implementation, the first determining module 501 determines M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, and determines the M energy The parameters each indicate a value relationship between the downmix signal energy and the residual signal energy of each of the M subbands, the M energy parameters correspond one-to-one with the M subbands, and M It is further configured to determine the energy parameter having the maximum value of the energy parameters as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by the first determination module 501 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] res_cod_NRG_M[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b.

Optionally, in one implementation, the first determination module 501 determines the sum of the M energy parameters as the first parameter res_dmx_ratio ₁ (to be corrected), res_dmx_ratio ₁ being the sum of the M energy parameters of which res_dmx_ratio_max and the downmix signal energy res_cod_NRG_M[b] of each of the M sub-bands, and determining res_dmx_ratio ₂ obtained by the correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, in one implementation, the res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

Optionally, in one implementation, the first determining module 501 bases the first parameter on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands. further configured to determine the

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, in one implementation, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the following equation:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, in one implementation, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the following equation:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by the first determination module 501 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. show.

Optionally, in one embodiment, the residual signal coding parameter of the current frame determined by the first determination module 501 is the long-term smoothing parameter of the current frame, and the long-term smoothing parameter of the current frame is The parameters satisfy the following formula,
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is the same as the preset second is greater than the value of α for greater than or equal to the threshold of , the second threshold is between 0 and 0.6 inclusive, the third threshold is between 2.7 and 3.7 inclusive, or the second parameter is prior to The value of α when the first parameter is greater than the preset fourth threshold, if greater than the preset fifth threshold is is greater than the value of α, the fourth threshold is between 0 and 0.9, the fifth threshold is between 0 and 0.71, or the first parameter is preset above the second threshold The value of α is small and the second parameter is greater than the preset third threshold, the second parameter is greater than or equal to the preset fifth threshold and is less than or equal to the preset third threshold , the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0.71 It is below.

Optionally, in one embodiment, the second determining module 502, when it is determined to encode the M subband residual signals, encodes the M subband downmix signals and the residual signals. It is further configured to encode the downmix signal of the M subbands when it is determined to encode or not to encode the residual signal of the M subbands.

FIG. 9 is a schematic block diagram of a stereo signal encoding device according to one embodiment of the present application. Apparatus 600 of FIG.
a memory 601 configured to store a program;
A processor 602 configured to execute a program stored in a memory 601, wherein when the program in the memory is executed, the processor 602 converts residual signal coding parameters of a current frame of a stereo signal into: determined based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame, wherein the residual signal coding parameters of the current frame encode the residual signals of the M subbands is used to indicate whether the M subbands are at least a portion of the N subbands, N is a positive integer greater than 1, M ≤ N, where N is a positive integer A processor 602, specifically configured to determine whether to encode the residual signal of the M subbands of the current frame based on the residual signal coding parameters.

Optionally, in one implementation, the M subbands are the M subbands whose subband index numbers are less than or equal to a preset maximum subband index number in the N subbands.

Optionally, in one embodiment, the M subbands have a subband index number greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number in the N subbands. are M subbands.

In one optional implementation, the processor 602 compares the residual signal coding parameter to a preset first threshold, and if the first threshold is greater than 0 and less than 1.0, the residual signal coding If the parameter is less than the first threshold, determine not to code the residual signal in each of the M subbands; or if the residual signal coding parameter is greater than the first threshold, the M subbands It is further configured to determine to encode the residual signal of each of the bands.

In an optional implementation, the processor 602 is further configured to determine residual signal coding parameters based on downmix signal energy, residual signal energy, and side gains for each of the M subbands. be.

In one optional implementation, processor 602 determines a first parameter based on the downmix signal energy, residual signal energy, and side gain for each of the M subbands, the first parameter being: indicating a value relationship between the downmix signal energy and the residual signal energy for each of the M subbands, a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands; , where the second parameter indicates a value relationship between the first energy sum and the second energy sum, and the first energy sum is the residual signal energy and the downmix signal energy of the M subbands and the second energy sum is the sum of the residual signal energies and the downmix signal energies of M subbands in the frequency domain signal of the previous frame of the current frame, and the M has the same subband index number as the M subbands of the previous frame, and the residual is calculated based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame. It is further configured to determine difference signal encoding parameters.

In one optional implementation, the processor 602 determines M energy parameters based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands; respectively denote the value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters correspond one-to-one with the M subbands and determining the energy parameter having the maximum value among the M energy parameters as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by processor 602 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/(res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g(b) is 4 represents a function of the side gain side_gain[b] of the subband whose subband index number is b.

In one optional implementation, the processor 602 determines the first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being the indicating a value relationship between the downmix signal energy and the residual signal energy for each of the bands, determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands; A second parameter indicates a value relationship between the first energy sum and the second energy sum, the first energy sum being the sum of the residual signal energy and the downmix signal energy of the M subbands. , the second energy sum is the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency domain signal of the previous frame of the current frame, and the M subbands of the current frame are Residual signal encoding based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame, with the same subband index number as the M subbands of the previous frame It is further configured to determine parameters.

In one optional implementation, the processor 602 determines M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, the M energy parameters being equal to the M and the residual signal energy of each of the subbands of the M energy parameters corresponding one-to-one with the M subbands, and the M energy parameters of It is further configured to determine the energy parameter having the maximum value thereof as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by processor 602 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] res_cod_NRG_M[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b.

In one optional implementation, the processor 602 determines the sum of the M energy parameters as the first parameter res_dmx_ratio ₁ (to be corrected), where res_dmx_ratio ₁ is the maximum value of the M energy parameters. It is further configured to correct based on the res_dmx_ratio_max and the downmix signal energy res_cod_NRG_M[b] of each of the M subbands and determine a res_dmx_ratio ₂ obtained by the correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, in one implementation, the res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

Optionally, in one implementation, the processor 602 determines the first parameter based on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands. further configured to

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, in one implementation, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the following equation:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, in one implementation, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the following equation:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by processor 602 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. show.

Optionally, in one implementation, the residual signal coding parameter determined by processor 602 is the long-term smoothing parameter of the current frame.

a residual signal coding parameter of the current frame determined by the processor 602 if the first parameter is less than a preset second threshold and the second parameter is greater than a preset third threshold; is the long-term smoothing parameter of the current frame, and the long-term smoothing parameter of the current frame satisfies the following equation,
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is the same as the preset second is greater than the value of α for greater than or equal to the threshold of , the second threshold is between 0 and 0.6 inclusive, the third threshold is between 2.7 and 3.7 inclusive, or the second parameter is prior to The value of α when the first parameter is greater than the preset fourth threshold, if greater than the preset fifth threshold is is greater than the value of α, the fourth threshold is between 0 and 0.9, the fifth threshold is between 0 and 0.71, or the first parameter is preset above the second threshold The value of α is small and the second parameter is greater than the preset third threshold, the second parameter is greater than or equal to the preset fifth threshold and is less than or equal to the preset third threshold , the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0.71 It is below.

Optionally, in one embodiment, the processor 602 encodes the M subband downmix and residual signals when it is determined to encode the M subband residual signals. , or is further configured to encode the downmix signal of the M subbands when it is determined not to encode the residual signal of the M subbands.

The present application further provides chips. The chip includes a processor and communication interface. The communication interface is configured to communicate with an external device, and the processor is configured to perform the stereo signal encoding method in the embodiments of the present application.

Optionally, in one implementation, the chip may further include memory. The memory stores instructions and the processor is configured to execute the instructions stored in the memory. When the instructions are executed, the processor is configured to perform the stereo signal encoding method in the embodiments of the present application.

Optionally, in one embodiment, the chip is incorporated into terminal equipment or network equipment.

The present application provides a computer-readable storage medium. The computer-readable storage medium stores program code to be executed by the device. The program code includes instructions for performing the stereo signal encoding method in the embodiments of the present application.

The processors referred to in the embodiments of the present invention may be Central Processing Units (CPUs), or other general purpose processors, Digital Signal Processors (DSPs), application specific integrated circuits. (Application Specific Integrated Circuit, ASIC), Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. . A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, and so on.

It will be appreciated that the memory referred to in embodiments of the present invention may be volatile memory or non-volatile memory, and may include volatile and non-volatile memory. Nonvolatile memory includes Read-Only Memory (ROM), Programmable Read-Only Memory (Programmable ROM, PROM), Erasable Programmable Read-Only Memory (Erasable PROM, EPROM), Electrically Erasable Programmable Read-Only Memory (Electrically EPROM, EEPROM), or flash memory. Volatile memory can be random access memory (RAM), used as an external cache. By way of example and not limitation, many forms of RAM such as Static Random Access Memory (Static RAM, SRAM), Dynamic Random Access Memory (Dynamic RAM, DRAM), Synchronous Dynamic Random Access Memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), sync Link Dynamic Random Access Memory (Synchlink DRAM, SLDRAM), Direct Rambus Random Access Memory (Direct Rambus RAM, DR RAM) may be used.

Note that if the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gates, transistor logic circuits, or discrete hardware components, memory (storage module) is integrated into the processor. sea bream.

Note that the memory described herein includes, but is not limited to, these memories and any other suitable types of memory.

Those skilled in the art will understand that each unit and algorithm step can be realized by electronic hardware or by a combination of computer software and electronic hardware in combination with the examples described in the embodiments disclosed herein. would do. Whether the function is performed by hardware or by software depends on the particular application and design constraints of the technical solution. Skilled artisans may implement the described functionality using varying methods for each particular application, but such implementations should not be considered beyond the scope of this application.

For convenience of explanation, the detailed operation processes of the aforementioned systems, devices, and units shall refer to the corresponding processes in the aforementioned method embodiments, and are not described in detail herein. This will be clearly understood by those skilled in the art.

It should be appreciated that in some of the embodiments provided in this application, the disclosed systems, devices, and methods may be implemented in other ways. For example, the described apparatus embodiment is merely exemplary. For example, the division into units is merely a logical functional division, and other divisions are possible in actual implementation. For example, multiple units or components may be combined or integrated into other systems, and some features may be ignored or not implemented. Additionally, the mutual couplings or direct couplings or communication connections shown or described may be implemented using some interfaces. Indirect couplings or communicative connections between devices or units may be realized in electronic, mechanical, or other form.

Units described as separate parts may or may not be physically separate and parts illustrated as units may or may not be physical units and may or may not be located together. may be distributed over multiple network units. Part or all of the units may be selected based on actual requirements to achieve the purpose of each embodiment's solution.

In addition, the functional units in the embodiments of the present application may be integrated into one processing unit, or each of the units may exist physically independently, or two or more units may be combined into one integrated into the unit.

When each function is implemented in the form of software functional units and sold or used as an independent product, the functions can be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the present application may be realized in the form of software products essentially or part of which contributes to the prior art, or part of the technical solutions. The computer software product is stored in a storage medium and instructs a computing device (which may be a personal computer, server, network appliance, etc.) to perform all or part of the method steps described in the embodiments of the present application. Including some commands to command. The aforementioned storage medium may store the program code, such as a USB flash drive, removable hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk. including any medium capable of

The above descriptions are only specific embodiments of the present application and are not intended to limit the protection scope of the present application. Any variation or replacement readily figured out by a person skilled in the art within the technical scope disclosed in the present application shall fall within the protection scope of the present application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

110 Encoding Components
120 decoding component
130 mobile terminals
131 collection components
132 channel coding components
140 mobile terminals
141 Audio Playback Components
142 channel decoding component
150 network elements
151 channel decoding component
152 channel coding components
300 stereo signal encoding method
500 devices
501 First Decision Module
502 second decision module
600 devices
601 memory
602 processor

This application is based on Chinese Patent Application No. 201810549237.3 entitled "STEREO SIGNAL ENCODING METHOD AND APPARATUS" filed with the Chinese Patent Office on May 31, 2018, which is incorporated herein by reference in its entirety. priority based on

The present application relates to the audio field, and more particularly to a stereo signal encoding method and stereo signal encoding apparatus.

The general process of encoding a stereo signal using time domain or time frequency domain stereo encoding techniques is as follows.
performing time-domain preprocessing on the left-channel time-domain signal and the right-channel time-domain signal;
performing time domain analysis on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing;
performing time frequency domain transformation on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing to obtain a left channel frequency domain signal and a right channel frequency domain signal;
determining the Inter-channel Time Difference (ITD) parameter in the time domain;
performing time shift adjustments to the left frequency domain signal and the right channel frequency domain signal based on the ITD parameters;
Compute stereo parameters, a downmix signal and a residual signal based on the left channel frequency domain signal and the right channel frequency domain signal obtained by time shift adjustment, and encode the stereo parameters, the downmix signal and the residual signal. .

In the prior art, only the stereo parameters and the downmix signal are generally coded when the coding rate is relatively low, and part or all of the residual signal is coded only when the coding rate is relatively high. It has been known. In this case, the spatial sensation of the decoded stereo signal is relatively low, and the sound image stability of the decoded stereo signal is relatively low.

In other prior art, it is known that when the coding rate is relatively low, in addition to the downmix signal, residual signals of subbands satisfying a preset bandwidth range are also coded. Although this coding method can improve the spatial perception and sound image stability of the decoded stereo signal, the total number of coding bits used for coding the residual signal and coding the downmix signal is Fixed, the low frequency information is preferentially encoded during downmix signal encoding, so that when the downmix signal is to be encoded, some signals are replaced with more abundant high frequencies in the downmix signal. There may not be enough bits to encode the information. Therefore, the high frequency distortion of the decoded stereo signal is relatively large, which affects the coding quality.

The present application aims to improve the spatial sensation and sound image stability of the decoded stereo signal and to reduce the high frequency distortion of the decoded stereo signal as much as possible, thereby improving the coding quality. , to provide a stereo signal encoding method.

According to a first aspect, a stereo signal encoding method is provided. The method comprises determining residual signal coding parameters for a current frame of a stereo signal based on downmix signal energy and residual signal energy for each of M subbands of the current frame, comprising: The residual signal coding parameter of the current frame is used to indicate whether to code the residual signal of the M subbands, where the M subbands are at least part of the N subbands. , where N is a positive integer greater than 1, M ≤ N, M is a positive integer, and the M subs of the current frame based on the residual signal coding parameters of the current frame and determining whether to encode the band's residual signal.

The residual signal coding parameters are determined based on the downmix signal energy and the residual signal energy of the M subbands within the N subbands, satisfying a preset bandwidth range; Whether to code the residual signal of each subband is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, the M subbands are the preset maximum subband index number among the N subbands. Here are the M subbands:

Optionally, in one embodiment, the M subbands have a subband index number greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number in the N subbands. are M subbands.

The minimum subband index number and/or maximum subband index number are set based on different coding rates. The residual signal coding parameters are determined based on the different coding rates and the downmix signal energy and residual signal energy of multiple specific subbands within the N subbands, and Whether to code each residual signal is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, the residual signal of each of the M subbands is encoded based on the residual signal coding parameters of the current frame. The step of determining whether the residual signal coding parameter of the current frame is compared to a preset first threshold, wherein the first threshold is greater than 0 and less than 1.0, the step and determining not to encode the residual signal of each of the M subbands if the residual signal coding parameter of the current frame is less than or equal to the first threshold; or and determining to encode the residual signal of each of the M subbands if the encoding parameter is greater than a first threshold.

A first threshold is set and the determined residual signal coding parameter is compared to the first threshold. Whether to encode the residual signal of each of the M subbands is determined based on the result of comparing the residual signal encoding parameter with the first threshold. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to encode all residual signals in subbands that satisfy a preset bandwidth range is determined based on the result of comparing the residual signal coding parameter with the first threshold. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, residual signal encoding of the current frame based on downmix signal energy and residual signal energy for each of the M subbands Determining the parameters includes determining residual signal coding parameters based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands.

A residual signal coding parameter for each of the M subbands is determined based on the downmix signal energy, the residual signal energy, and the side gain to encode the residual signal for each of the M subbands. is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Referring to the first aspect, in one possible implementation of the first aspect, residual signal encoding based on downmix signal energy, residual signal energy, and side gain for each of the M subbands Determining the parameters includes determining a first parameter based on the downmix signal energy, residual signal energy, and side gain for each of the M subbands, wherein the first parameter is M second step based on the downmix signal energy and the residual signal energy for each of the M subbands, indicating a value relationship between the downmix signal energy and the residual signal energy for each of the M subbands; wherein the second parameter indicates a value relationship between the first energy sum and the second energy sum, and the first energy sum is the residual signal of the M subbands a sum of the energy and the downmix signal energy, a second energy sum being the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency-domain signal of the previous frame of the current frame; a step in which the M subbands of the current frame have the same subband index numbers as the M subbands of the previous frame; determining residual signal coding parameters for the current frame based on the long-term smoothing parameters.

Referring to the first aspect, in one possible implementation of the first aspect, the first parameter is determined based on the downmix signal energy, the residual signal energy, and the side gain of each of the M subbands. The determining step is determining M energy parameters based on the downmix signal energy, residual signal energy, and side gain for each of the M subbands, wherein the M energy parameters are equal to M step, respectively showing a value relationship between the downmix signal energy and the residual signal energy of each of the subbands, the M energy parameters corresponding one-to-one with the M subbands; and determining as the first parameter the energy parameter having the maximum value among the energy parameters of .

Referring to the first aspect, in one possible implementation of the first aspect, the energy parameter of the subband with subband index number b among the M energy parameters satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/(res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g(b) is 4 represents a function of the side gain side_gain[b] of the subband whose subband index number is b.

Referring to the first aspect, in one possible implementation of the first aspect, residual signal encoding of the current frame based on downmix signal energy and residual signal energy for each of the M subbands Determining the parameter is determining a first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being the and determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands. wherein the second parameter indicates a value relationship between the first energy sum and the second energy sum, the first energy sum being the residual signal energy of the M subbands and the downmix is the sum of the signal energies, the second energy sum is the sum of the residual signal energies and the downmix signal energies of the M subbands in the frequency domain signal of the frame before the current frame, and the sum of the downmix signal energies of the current frame; A step in which the M subbands have the same subband index number as the M subbands of the previous frame, and the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame. determining the residual signal coding parameters for the current frame based on .

Referring to the first aspect, in one possible implementation of the first aspect, determining the first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands comprises: , determining M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, wherein the M energy parameters are based on the downmix signal energy and the residual signal energy for each of the M subbands. Steps and maximum values among the M energy parameters, each showing a value relationship between the mix signal energy and the residual signal energy, wherein the M energy parameters correspond one-to-one with the M subbands; as the first parameter.

Optionally, in one implementation, the sum of the M energy parameters is determined as the first parameter res_dmx_ratio ₁ (to be corrected), where res_dmx_ratio ₁ is the maximum of the M energy parameters res_dmx_ratio_max and Corrected based on the downmix signal energy res_cod_NRG_M[b] of each of the M subbands to determine res_dmx_ratio ₂ obtained by correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, in one implementation, the res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

Optionally, in one embodiment, the encoder side determines the first parameter based on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands.

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, in one implementation, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the following equation:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, in one implementation, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the following equation:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr

Referring to the first aspect, in one possible implementation of the first aspect, the energy parameter of the subband with subband index number b among the M energy parameters satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] res_cod_NRG_M[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b.

Referring to the first aspect, in one possible implementation of the first aspect, the current frame residual signal coding parameter is the current frame long-term smoothing parameter, and the current frame long-term smoothing parameter parameter satisfies the following equation,
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is the same as the preset second is greater than the value of α for greater than or equal to the threshold of , the second threshold is between 0 and 0.6 inclusive, the third threshold is between 2.7 and 3.7 inclusive, or the second parameter is prior to The value of α when the first parameter is greater than the preset fourth threshold is less than the preset fifth threshold, and the value of α when the first parameter is less than or equal to the preset fourth threshold. greater than the value of α and the fourth threshold is between 0 and 0.9 and the fifth threshold is between 0 and 0.71 or the second parameter is greater than or equal to the preset fifth threshold and is less than or equal to a preset third threshold, then the value of α is less than the first parameter preset second threshold and the second parameter is preset third threshold less than the value of α if greater than, the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0.71 It is below.

Referring to the first aspect, in one possible implementation of the first aspect, when it is determined to encode the residual signal of M subbands, the method comprises: encoding the downmix signal and the residual signal, or encoding the downmix signal for the M subbands when it is determined not to encode the residual signal for the M subbands; Including further.

According to a second aspect, an encoding device is provided. The apparatus is configured to determine residual signal coding parameters for a current frame of the stereo signal based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame. A first determining module, wherein the residual signal coding parameters of the current frame are used to indicate whether to code the residual signal of M subbands, wherein the M subbands are N a first determining module, wherein N is a positive integer greater than 1 and M≤N, where M is a positive integer; and residual signal encoding of the current frame. and a second decision module configured to decide whether to encode the M subband residual signals based on the parameters.

According to a third aspect, an encoding device is provided. The apparatus includes a memory and a processor, the memory configured to store the program, the processor configured to execute the program, and when the program is executed, the processor performs the first aspect or the first aspect. A method according to any one of the possible embodiments of aspect 1 is carried out.

According to a fourth aspect, a computer-readable storage medium is provided. The computer-readable storage medium stores program code to be executed by the device, the program code including instructions used to perform the method according to the first aspect or various implementations of the first aspect.

According to a fifth aspect, a chip is provided. The chip includes a processor and communication interface. A communication interface is configured to communicate with an external device. The processor is configured to perform the method according to the first aspect or any one of the possible implementations of the first aspect.

Optionally, in one implementation, the chip may further include memory. The memory stores instructions and the processor is configured to execute the instructions stored in the memory. When the instructions are executed, the processor is configured to perform the method according to the first aspect or any one of the possible implementations of the first aspect.

Optionally, in one embodiment, the chip is incorporated into terminal equipment or network equipment.

1 is a schematic structural diagram of stereo encoding and decoding in the time domain according to an embodiment of the present application; FIG. 1 is a schematic diagram of a mobile terminal according to an embodiment of the present application; FIG. 1 is a schematic diagram of a network element according to an embodiment of the application; FIG. 1 is a schematic flow diagram of a stereo signal encoding method in the frequency domain; 1 is a schematic flow diagram of a stereo signal encoding method in the time-frequency domain; 1 is a schematic flow diagram of a stereo signal encoding method according to an embodiment of the present application; Fig. 4 is another schematic flow diagram of a stereo signal encoding method according to an embodiment of the present application; 1 is a schematic block diagram of a stereo signal encoding device according to an embodiment of the present application; FIG. Fig. 3 is another schematic block diagram of a stereo signal encoding device according to an embodiment of the present application;

The technical solutions of the present application are described below with reference to the accompanying drawings.

FIG. 1 is a schematic structural diagram of a stereo encoding and decoding system in the time domain according to an exemplary embodiment of the present application. Stereo encoding and decoding system includes encoding component 110 and decoding component 120 .

Encoding component 110 is configured to encode the stereo signal in the time domain. Optionally, encoding component 110 may be implemented using software, or may be implemented using hardware, or may be implemented in the form of a combination of software and hardware. . This is not limited in this embodiment.

Encoding component 110 encodes the stereo signal in the time domain and includes the following steps.

(1) Performing time domain preprocessing on the obtained stereo signal to obtain a left channel signal obtained by the time domain preprocessing and a right channel signal obtained by the time domain preprocessing.

A stereo signal is collected by the collection component and sent to the encoding component 110 . Optionally, the collection component and encoding component 110 may be located on the same device. Alternatively, the collection component and encoding component 110 may be located on different devices.

The left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing are signals of two channels of the stereo signal obtained by preprocessing.

Optionally, pre-processing includes at least one of high-pass filtering, pre-emphasis, sampling rate conversion, and channel conversion. This is not limited in this embodiment.

(2) Perform delay estimation based on the left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing, and calculate the difference between the left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing. Get the inter-channel time difference.

(3) performing delay adjustment processing on the left channel signal obtained by preprocessing and the right channel signal obtained by preprocessing based on the inter-channel time difference, and obtaining the left channel signal and delay matching processing obtained by delay matching processing; Obtain the right channel signal obtained by

(4) encoding the inter-channel time difference to obtain an encoding index of the inter-channel time difference;

(5) calculating the stereo parameters used in the time-domain downmixing process, encoding the stereo parameters used in the time-domain downmixing process, and obtaining the encoding index of the stereo parameters used in the time-domain downmixing process as obtain.

The stereo parameters used for time-domain downmix processing are used to perform time-domain downmix processing on the left channel signal obtained by delay matching processing and the right channel signal obtained by delay matching processing.

(6) performing time-domain downmix processing on the left channel signal obtained by delay matching processing and the right channel signal obtained by delay matching processing, based on the stereo parameters used in the time domain downmix processing, to obtain a primary Obtain a channel signal and a secondary channel signal.

A primary channel signal is used to represent information about the correlation between channels. Secondary channel signals are used for information about differences between channels. The secondary channel signal is minimal when the left channel signal obtained by the delay matching process and the right channel signal obtained by the delay matching process are matched in the time domain. In this case the stereo signal has the best effect.

(7) separately encode the primary channel signal and the secondary channel signal to obtain a first mono-encoded bitstream corresponding to the primary channel signal and a second mono-encoded bitstream corresponding to the secondary channel signal; .

(8) Write the encoding index of the inter-channel time difference, the encoding index of the stereo parameter, the first mono-encoded bitstream, and the second mono-encoded bitstream into the stereo-encoded bitstream.

Decoding component 120 is configured to decode the stereo-encoded bitstream produced by encoding component 110 to obtain a stereo signal.

Optionally, encoding component 110 is wired or wirelessly connected to decoding component 120, over which decoding component 120 obtains the stereo-encoded bitstream produced by encoding component 110. . Alternatively, encoding component 110 stores the generated stereo-encoded bitstream in memory, and decoding component 120 reads the stereo-encoded bitstream in memory.

Optionally, decoding component 120 may be implemented using software, or may be implemented using hardware, or may be implemented in the form of a combination of software and hardware. This is not limited in this embodiment.

Decoding component 120 decodes the stereo-encoded bitstream to obtain a stereo signal, which includes the following steps.

(1) Decoding the first mono-encoded bitstream and the second mono-encoded bitstream in the stereo-encoded bitstream to obtain a primary channel signal and a secondary channel signal.

(2) based on the stereo-encoded bitstream, obtain the coding indices of the stereo parameters used for the time-domain upmixing process, perform the time-domain upmixing process on the primary channel signal and the secondary channel signal, and obtain the time domain upmixing process; A left channel signal obtained by domain upmix processing and a right channel signal obtained by time domain upmix processing are obtained.

(3) obtaining a coding index for the inter-channel time difference based on the stereo coded bitstream, and performing delay adjustment on the left channel signal obtained by the time domain upmixing process and the right channel signal obtained by the time domain upmixing process; Go and get a stereo signal.

Optionally, encoding component 110 and decoding component 120 may be located on the same device or may be located on different devices. The device can be a mobile terminal with audio signal processing capabilities, such as a mobile phone, tablet computer, laptop portable computer, desktop computer, Bluetooth® speaker, pen recorder, or wearable device, or a core network or It may be a network element with audio signal processing capabilities in a wireless network. This is not limited in this embodiment.

For example, as shown in FIG. 2, in this embodiment encoding component 110 is located at mobile terminal 130, decoding component 120 is located at mobile terminal 140, and mobile terminal 130 and mobile terminal 140 are: Mutually independent devices with audio signal processing capabilities, such as mobile phones, wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, etc. Mobile terminals Description is made using an example where 130 is connected to mobile terminal 140 using a wireless or wired network.

Optionally, mobile terminal 130 includes collection component 131 , encoding component 110 and channel encoding component 132 . Collection component 131 is connected to encoding component 110 and encoding component 110 is connected to channel encoding component 132 .

Optionally, mobile terminal 140 includes audio reproduction component 141 , decoding component 120 and channel decoding component 142 . The audio reproduction component 141 is connected to the decoding component 120 and the decoding component 120 is connected to the channel decoding component 142 .

After acquiring the stereo signal using the acquisition component 131, the mobile terminal 130 encodes the stereo signal using the encoding component 110 to obtain a stereo encoded bitstream, and then a channel encoding configuration. Encode the stereo-encoded bitstream using element 132 to obtain the transmitted signal.

Mobile terminal 130 transmits transmission signals to mobile terminal 140 using a wireless or wired network.

After receiving the transmitted signal, mobile terminal 140 decodes the transmitted signal using channel decoding component 142 to obtain a stereo-encoded bitstream, and decodes the stereo-encoded bitstream using decoding component 120 . Decode to obtain a stereo signal and reproduce the stereo signal using audio reproduction component 141 .

For example, as shown in FIG. 3, this embodiment provides an example in which encoding component 110 and decoding component 120 are located in network element 150 with audio signal processing capability within the same core network or wireless network. is used to explain.

Optionally, network element 150 includes channel decoding component 151 , decoding component 120 , encoding component 110 and channel encoding component 152 . Channel decoding component 151 is connected to decoding component 120 , decoding component 120 is connected to encoding component 110 , and encoding component 110 is connected to channel encoding component 152 .

After receiving a transmission transmitted by another device, channel decoding component 151 decodes the transmission to obtain a first stereo-encoded bitstream, and decoding component 120 decodes the first stereo-encoded bitstream. to obtain a stereo signal, encoding component 110 encodes the stereo signal to obtain a second stereo-encoded bitstream, and channel encoding component 152 encodes the second stereo-encoded bitstream. to obtain the transmitted signal.

The other device may be a mobile terminal with audio signal processing capability or other network element with audio signal processing capability. This is not limited in this embodiment.

Optionally, encoding component 110 and decoding component 120 within the network element may transcode the stereo-encoded bitstream transmitted by the mobile terminal.

Optionally, in this embodiment, the device in which encoding component 110 is installed is referred to as an audio encoding device. In actual implementation, the audio encoding device may also have audio decoding functionality. This is not limited in this embodiment.

Optionally, this embodiment is described using only stereo signals as an example. In the present application, the audio encoding device may further process a multi-channel signal, the multi-channel signal comprising signals of at least two channels.

In order to facilitate understanding of the stereo signal encoding method in the embodiments of the present application, the following will first refer to FIG. 4 and FIG. The entire encoding process of the method is generally described.

FIG. 4 is a schematic flow diagram of a stereo signal encoding method in the frequency domain. This encoding method specifically includes 101-107.

101: Convert the time-domain stereo signal into a frequency-domain stereo signal.

102: Extract the frequency domain stereo parameters in the frequency domain.

103: Downmix processing is performed on the stereo signal in the frequency domain to obtain a downmix signal and a residual signal.

A downmix signal may also be referred to as a central channel signal or primary channel signal, and a residual signal may be referred to as a side channel signal or secondary channel signal.

104: Encoding the downmix signal to obtain encoding parameters corresponding to the downmix signal, and writing the encoding parameters into the encoded bitstream.

106: Encode the stereo parameters in the frequency domain to obtain encoding parameters corresponding to the stereo parameters in the frequency domain, and write the encoding parameters into the encoded bitstream.

In an optional implementation, the method may further include 105: encoding the residual signal to obtain encoding parameters corresponding to the residual signal, and writing the encoding parameters to the encoded bitstream.

107: Multiplex the bitstream.

FIG. 5 is a schematic flow chart of a stereo signal encoding method in the time-frequency domain. This encoding method specifically includes 201-208.

201: Perform time domain analysis on the stereo signal and extract time domain stereo parameters.

202: Transform the time-domain stereo signal into a frequency-domain stereo signal.

203: Extract the frequency domain stereo parameters in the frequency domain.

204: Perform downmix processing on the stereo signal in the frequency domain to obtain a downmix signal and a residual signal.

205: Encoding the downmix signal to obtain encoding parameters corresponding to the downmix signal, and writing the encoding parameters into the encoded bitstream.

207: Encoding the time-domain stereo parameters and the frequency-domain stereo parameters to obtain encoding parameters corresponding to the time-domain stereo parameters and encoding parameters corresponding to the frequency-domain stereo parameters, and encoding the encoding parameters Write to bitstream.

Optionally, the method further comprises 206: encoding the residual signal to obtain encoding parameters corresponding to the residual signal, and writing the encoding parameters to the encoded bitstream.

208: Multiplex the bitstream.

When the coding rate is relatively low, for example, when the coding bandwidth is Wideband, the coding rate is relatively low, Kilo-bytes per second (kbps), 16.4 kbps, 24. When it is 4kbps, or 32kbps, etc., the downmix signal of each frame of the stereo signal is coded to improve the sense of space and stability during playback of the stereo signal and to reduce the high frequency distortion of the stereo signal. Then, all residual signals in subbands satisfying a preset bandwidth range are coded. Alternatively, if the coding rate is relatively low, only the stereo parameters and the downmix signal are coded. Part or all of the residual signal is encoded only if the encoding rate is relatively high, such as 48 kbps, 64 kbps, or 96 kbps. The present application provides a stereo signal encoding method. In this way, the spatial perception and sound image stability of the decoded stereo signal can be improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the overall coding quality. be done.

FIG. 6 is a schematic flow diagram of a stereo signal encoding method 300 according to one embodiment of the present application. Method 300 may be performed by an encoder side, which may be an encoder or a device with stereo signal encoding capabilities. Method 300 includes the following steps.

The stereo signal encoding method of the present application may be a stereo encoding method that can be applied independently, or a stereo encoding method that is applied to multi-channel signal encoding. The encoder side processes the stereo signal frame by frame. In the following, using a wideband stereo signal where the signal length of each frame is 20ms as an example, and using the frame being processed by the encoder side (e.g. the current frame) as an example, the stereo signal encoding of the method 300 A detailed description will be given of the conversion method.

301: Determine a residual signal coding parameter of the current frame of the stereo signal based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame, and determine the residual signal of the current frame A signal coding parameter is used to indicate whether to code the residual signal of M subbands, where M subbands are at least part of the N subbands and N is greater than 1. A large positive integer, M≤N, where M is a positive integer.

Specifically, the encoder side divides the spectral coefficients of the current frame of the stereo signal to obtain N subbands, and divides the spectral coefficients of the current frame of the stereo signal into at least some of the N subbands (e.g., based on the downmix signal energy and the residual signal energy of each of the M subbands of M ≤ N), the encoder side determines the residual signal coding parameters of the current frame A signal coding parameter may be used to determine whether to code the residual signal for each of the M subbands.

302: Determine whether to code the residual signals of the M subbands of the current frame according to the residual signal coding parameters of the current frame.

Specifically, the encoder side encodes the residual signal of each of the M subbands of the current frame based on the residual signal coding parameters of the current frame determined in step 301. to decide whether

Upon determining to encode the residual signal of each of the M subbands, the downmix signal and residual signal of each of the M subbands are encoded.

If it is determined not to encode the residual signal for each of the M subbands, then the downmix signal for each of the M subbands is encoded.

In one implementation, by way of example and not limitation, the M subbands are the M subbands within the N subbands whose subband index numbers are less than a preset maximum subband index number. In other words, the M subbands are subbands with relatively low frequencies within the N subbands, specifically, the frequencies of the M subbands are equal to M lower than the frequencies of the NM subbands other than the subbands.

Specifically, since different maximum subband index numbers are preconfigured based on different coding rates, M subbands whose subband index numbers are less than or equal to the preconfigured maximum subband index number are preconfigured. One of the N subbands is selected based on the configured maximum subband index number, and the residual signal coding parameters of the current frame are determined based on the M subbands.

For example, if the coding rate is 26 kbps, N=10, M=5, and the preset maximum subband index number is set to 4, this means that the residual signal coding parameters of the current frame are It is determined based on 5 subbands with subband index numbers 0 to 4 in 10 subbands.

In another example, if the coding rate is 44 kbps, N=12, M=6, and the preset maximum subband index number is set to 5, this is the residual signal coding for the current frame. It shows that the parameters are determined based on 6 subbands with subband index numbers 0 to 5 in 12 subbands.

In another example, if the coding rate is 56 kbps, N=12, M=7, and the preset maximum subband index number is set to 6, this is the residual signal coding for the current frame. It shows that the parameters are determined based on 7 subbands with subband index numbers 0 to 6 in 12 subbands.

In other implementations, for different coding rates, the maximum subband index number and the minimum subband index number of the M subbands at different coding rates can be preset, so the subband index numbers are preset N subbands that are greater than or equal to the minimum subband index number and less than or equal to the preset maximum subband index number are N based on the preset maximum subband index number and the preset minimum subband index number , and the residual signal coding parameters for the current frame are determined based on the M subbands.

For example, if the coding rate is 26 kbps, N=10, M=4, the preset minimum subband index number is set to 4, and the preset maximum subband index number is set to 7, This indicates that the residual signal coding parameters of the current frame are determined based on four subbands with subband index numbers 4 to 7 among the ten subbands.

By way of example and not limitation, determining whether to encode the residual signal of each of the M subbands based on the residual signal coding parameters of the current frame comprises: determining whether to encode the residual signal of based on the result of comparing the residual signal coding parameter of the current frame with a preset first threshold, wherein the first threshold is A step greater than 0 and less than 1.0, and determining not to encode the residual signal of each of the M subbands if the residual signal coding parameter of the current frame is less than or equal to the first threshold. or determining to encode the residual signal of each of the M subbands if the residual signal coding parameter is greater than a first threshold.

Specifically, the encoder side compares the residual signal coding parameter of the current frame with a preset first threshold, and the residual signal coding parameter of the current frame is greater than the first threshold. If so, determine to encode the residual signal of each of the M subbands, or if the residual signal encoding parameter of the current frame is less than or equal to the first threshold, determine to encode the residual signal of each of the M subbands. Decide not to encode the residual signal.

For example, in one implementation, the first threshold is 0.075. If the value of the residual signal coding parameter of the current frame is 0.06, the encoder side does not code the residual signal of each of the M subbands.

It should be appreciated that the value of the first threshold is only an example and that the first threshold may alternatively be other values greater than 0 and less than 1.0. For example, the first threshold is 0.55, 0.46, 0.86, or 0.9.

In another optional implementation, the encoder side may use 0 or 1 to further indicate the comparison result between the residual signal coding parameter of the current frame and the first threshold. For example, 0 is used to indicate that the residual signal of each of the M subbands should not be coded, and 1 is used to indicate that the residual signal of each of the M subbands is coded. Used to indicate that something should be done. Of course, 1 may alternatively be used to indicate that the residual signal in each of the M subbands should not be encoded, and 0 alternatively may be used to indicate that each residual signal of is to be encoded.

In the following, in order to explain in detail how the encoder side determines the residual signal coding parameters of the current frame, M subbands are assigned the maximum subband index number ( For example, the maximum subband index number is M−1).

Method 1

The encoder side determines the residual signal coding parameters of the current frame based on the downmix signal energy, residual signal energy and side gain of each of the M subbands.

In one possible implementation, the encoder side determines the first parameter based on the downmix signal energy, the residual signal energy, and the side gain for each of the M subbands, the first parameter being , denoting the value relationship between the downmix signal energy and the residual signal energy for each of the M subbands,
determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the second parameter being a value relationship between the first energy sum and the second energy sum where the first energy sum is the sum of the residual signal energies and the downmix signal energies of the M subbands, and the second energy sum is the M is the sum of the residual signal energy and the downmix signal energy of the subbands of , where the M subbands of the current frame have the same subband index numbers as the M subbands of the previous frame, and
A residual signal coding parameter of the current frame is finally determined based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame.

Specifically, when determining the first parameter, the encoder side determines M energy parameters based on the downmix signal energy, residual signal energy, and side gain of each of the M subbands. and M energy parameters each indicating a value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters representing the M subbands. In one-to-one correspondence with the band, the encoder side finally determines the energy parameter having the maximum value among the M energy parameters as the first parameter.

Optionally, the energy parameter of the subband with subband index number b among the M energy parameters may be determined using the function:
res_dmx_ratio[b] = f(g(b), res_cod_NRG_M[b], res_cod_NRG_S[b]) (1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. where g(b) represents the function of the side gain side_gain[b] of the subband with subband index number b.

Specifically, in one embodiment, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies the following equation.
res_dmx_ratio[b] = res_cod_NRG_S[b] / (res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b] + 1) (2)

The first parameter is denoted res_dmx_ratio, where res_dmx_ratio satisfies the following equation.
res_dmx_ratio = max(res_dmx_ratio[0], res_dmx_ratio[1], ..., res_dmx_ratio[M-1]) (3)

When determining the second parameter, the encoder side first separately determines the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands; Let dmx_nrg_all_curr be the sum of the downmix signals of the subbands, and res_nrg_all_curr be the sum of the residual signal energies of the M subbands.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the formula:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

It should be appreciated that the value of γ ₁ is only an example and that the value of γ ₁ may alternatively be other values between 0 and 1 inclusive. For example, γ ₁ is 0.3, 0.5, 0.6, or 0.8.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the formula:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

It should be appreciated that the value of γ ₂ is only an example and that the value of γ ₂ may alternatively be other values between 0 and 1 inclusive. For example, _γ2 is 0.2, 0.5, 0.7, or 0.9.

The encoder side determines the sum of the downmix signal energy and the residual signal energy (ie, the first energy sum) of the M subbands of the current frame based on dmx_nrg_all_curr and res_nrg_all_curr. The first energy sum is denoted dmx_res_all.

Optionally, dmx_res_all satisfies the following formula.
dmx_res_all = res_nrg_all_curr + dmx_nrg_all_curr(6)

The encoder side may further determine the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency domain signal of the frame previous to the current frame (i.e., the second energy sum). Well, the M subbands of the frame before the current frame have the same subband index numbers as the M subbands of the current frame . The second energy sum is denoted dmx_res_all_prev.

For the determination of the second energy sum dmx_res_all_prev, please refer to the method for determining the first energy sum dmx_res_all described above. For the sake of brevity, the details are not repeated here.

After determining the first energy sum and the second energy sum, the encoder side may determine a second parameter based on the first energy sum and the second energy sum.

Optionally, the second parameter is the frame-to-frame energy variation rate, and the frame-to-frame energy variation rate is denoted as frame_nrg_ratio.

Optionally, in one implementation, the frame-to-frame energy variation rate frame_nrg_ratio satisfies the following equation.
frame_nrg_ratio = dmx_res_all/dmx_res_all_prev(7)

Optionally, in other implementations, the frame-to-frame energy variation rate frame_nrg_ratio satisfies the following equation.
frame_nrg_ratio = min(5.0, max(0.2, dmx_res_all/dmx_res_all_prev)) (8)

The max function is used to return the larger value at the given parameters (0.2, frame_nrg_ratio_prev) and the min function is used to return the larger value at the given parameters (5.0, max(0.2, frame_nrg_ratio_prev)) Used to return the minimum value. Compared to Eq. (7), Eq. (8) has an additional correction operation, so the frame_nrg_ratio determined using Eq. (8) represents the interframe energy variation between the current frame and the previous frame. can be better reflected.

After determining the first parameter and the second parameter, the encoder side calculates the residual of the current frame based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame. Difference signal encoding parameters may be determined.

By way of example and not limitation, the residual signal coding parameters for the current frame may be long-term smoothing parameters for the current frame. In other words, the encoder side determines the long-term smoothing parameter of the current frame based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame, and then M A long-term smoothing parameter of the current frame may be compared to a preset first threshold to determine whether to code the residual signal of each of the subbands of .

For example, the long-term smoothing parameter for the current frame satisfies
res_dmx_ratio_lt = res_dmx_ratio α + res_dmx_ratio_lt_prev (1 - α) (9)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 is.

When res_dmx_ratio_lt is calculated according to equation (9), if the value of the first parameter and/or the value of the second parameter changes, the value of parameter α in equation (9) may change accordingly. In other words, when the value of the first parameter and/or the value of the second parameter change, the weight of the long-term smoothing parameter of the frame before the current frame in equation (9) may change accordingly. .

For example, if the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is equal to the preset value of the first parameter. greater than the value of α for greater than or equal to the second threshold, where the second threshold is between 0 and 0.6, and the third threshold is between 2.7 and 3.7, or the second parameter is less than a preset fifth threshold, the value of α if the first parameter is greater than a preset fourth threshold is less than or equal to the preset fourth threshold the fourth threshold is between 0 and 0.9 inclusive, the fifth threshold is between 0 and 0.71 inclusive, or the second threshold with the first parameter preset If the value of α is less than the threshold and the second parameter is greater than the preset third threshold, the second parameter is greater than or equal to the preset fifth threshold and the preset third is greater than the value of α when it is less than or equal to the threshold, the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0 .71 or less.

For example, the value of the second threshold may be 0.1 and the value of the third threshold may be 3.2. Specifically, if the second parameter frame_nrg_ratio is greater than 3.2, the The value of α when the parameter res_dmx_ratio of 1 is less than 0.1 is greater than the value of α when res_dmx_ratio is greater than or equal to 0.1, or the value of the fourth threshold is 0.4 and the value of the fifth threshold may be 0.21. Specifically, when frame_nrg_ratio is less than 0.21, the value of α when res_dmx_ratio is greater than 0.4 is the value of α when res_dmx_ratio is 0.4 or less. or the second threshold value is 0.1, the third threshold value is 3.2, and the fifth threshold value is 0.21, specifically the value of α when res_dmx_ratio is less than 0.1 and frame_nrg_ratio is greater than 3.2 is greater than the value of α when frame_nrg_ratio is between 0.21 and 3.2, or the fourth threshold may be 0.4 and the value of the fifth threshold may be 0.21, specifically, the value of α when res_dmx_ratio is greater than 0.4 and frame_nrg_ratio is less than 0.21 is greater than the value of α when frame_nrg_ratio is 0.21 or more and 3.2 or less.

Further, for example, when res_dmx_ratio is less than 0.1 and frame_nrg_ratio is greater than 3.2, the value of α is 0.5, or when frame_nrg_ratio is between 0.21 and 3.2, the value of α is 0.1.

Note that the stated second through fifth threshold values and α values are illustrative examples only and do not constitute any limitations on the present application. The second through fifth threshold values and the value of α may alternatively be other values in a given interval.

Note further that the current frame has no previous frame if it is the first frame processed by the encoder side. In this case, when the long-term smoothing parameter of the current frame is determined, the long-term smoothing parameter of the frame before the current frame in the above formula is the preset long-term smoothing parameter. By way of example and not limitation, the value of the preset long-term smoothing parameter may be 1.0, or of course other values such as 0.9 or 1.1.

Method 2

The method for determining the residual signal coding parameter in Method 2 is similar to the method in Method 1, the difference being that the method for determining the first parameter is different. Reference may be made to the related discussion of determining signal coding parameters. For the sake of brevity, only the method for determining the first parameter in Method 2 is described herein.

By way of example and not limitation, the encoder side determines a first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being the 2 shows the value relationship between the downmix signal energy and the residual signal energy for each of .

Specifically, when determining the first parameter, the encoder side determines M energy parameters based on the downmix signal energy and the residual signal energy of each of the M subbands; energy parameters each indicate a value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters are paired with the M subbands. 1, and the encoder side finally determines the energy parameter having the maximum value among the M energy parameters as the first parameter.

Optionally, the energy parameter of the subband whose subband index number is b among the M energy parameters determined by the encoder side may be determined using the following function:
res_dmx_ratio[b] = f(res_cod_NRG_M[b], res_cod_NRG_S[b]) (10)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. show.

For example, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies the following equation.
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b] (11)

The first parameter is denoted res_dmx_ratio, where res_dmx_ratio satisfies the following equation.
res_dmx_ratio = max(res_dmx_ratio[0], res_dmx_ratio[1], ..., res_dmx_ratio[M-1]) (12)

After determining the first parameter, the encoder side determines the second parameter according to the method described in Method 1, finally determines the residual signal coding parameter according to the method described in Method 1, and Based on the difference signal coding parameters, it may be determined whether to code the residual signal for each of the M subbands.

Method 3

The method for determining the residual signal coding parameter in method 3 is similar to the method in method 1, the difference being that the method for determining the first parameter is different. Reference may be made to the related discussion of determining signal coding parameters. For the sake of brevity, only the method for determining the first parameter in Method 3 is described herein.

By way of example and not limitation, the encoder side determines a first parameter based on downmix signal energy and residual signal energy for each of the M subbands, corrects the first parameter, and final As the first parameter, determine the first parameter obtained by the correction, the first parameter indicating the value relationship between the downmix signal energy and the residual signal energy for each of the M subbands .

Specifically, when determining the first parameter, the encoder side determines M energy parameters based on the downmix signal energy and the residual signal energy of each of the M subbands; energy parameters each indicate a value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters are paired with the M subbands. 1, and the encoder side determines the sum of M energy parameters as the first parameter.

Optionally, the energy parameter of the subband whose subband index number is b among the M energy parameters determined by the encoder side may be determined using function (1).

For example, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies Equation (2).

Optionally, the energy parameter of the subband whose subband index number is b among the M energy parameters determined by the encoder side may be determined using function (11).

For example, among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies Equation (11).

For example, the first parameter res_dmx_ratio ₁ determined by the encoder side based on M energy parameters satisfies the following equation.

In addition, the encoder side may further determine the maximum value res_dmx_ratio_max among the M energy parameters, where res_dmx_ratio_max satisfies Equation (12).

The encoder side corrects res_dmx_ratio ₁ based on res_dmx_ratio_max of each of the M subbands and downmix signal energy res_cod_NRG_M[b], and determines res_dmx_ratio ₂ obtained by the correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2) (15)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

After determining the first parameter obtained by correction (res_dmx_ratio ₃ obtained by correction), the encoder side determines the second parameter according to the method described in Method 1, and determines the second parameter according to the method described in Method 1. A residual signal coding parameter may be finally determined, and whether to code the residual signal for each of the M subbands may be determined based on the residual signal coding parameter.

Method 4

The method for determining the residual signal coding parameter in method 4 is similar to the method in method 1, the difference being that the method for determining the first parameter is different. Reference may be made to the related discussion of determining signal coding parameters. For the sake of brevity, only the method for determining the first parameter in method 4 is described herein.

By way of example and not limitation, the encoder side determines the first parameter based on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands.

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

Optionally, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies equation (4) .

Optionally, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies equation (5) .

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr(16)

After determining the first parameter, the encoder side determines the second parameter according to the method described in Method 1, finally determines the residual signal coding parameter according to the method described in Method 1, and Based on the difference signal coding parameters, it may be determined whether to code the residual signal for each of the M subbands.

In order to better understand the whole encoding of stereo signals, in the following, we will use a wideband stereo signal as an example where the signal length of each frame is 20ms, and the frame being processed by the encoder side (e.g. the current frame ) as an example, the stereo signal encoding method 300 of this embodiment of the present application will be described with reference to FIG. The stereo signal encoding method shown in FIG. 7 includes at least the following steps.

401: Perform time domain preprocessing on the left channel time domain signal and the right channel time domain signal to obtain a left channel time domain signal and a right channel time domain signal obtained by the time domain preprocessing.

Specifically, the signal length of the current frame is 20ms. If the sampling frequency is 16 kHz ( kHz ), after sampling, the frame length of the current frame H=320, in other words the current frame contains 320 sampling points.

The current frame stereo signal includes a current frame left channel time domain signal and a current frame right channel time domain signal. The left channel time domain signal of the current frame is denoted as x _L (n) and the right channel time domain signal of the current frame is denoted as x _R (n). n is the sequence number of sampling points, n=0, 1, . . . , and H−1. Left channel time domain signals and right channel time domain signals may be referred to as left and right channel time domain signals.

The step of performing time-domain preprocessing on the left-channel time-domain signal and the right-channel time-domain signal of the current frame includes converting the left-channel time-domain signal and the right-channel time-domain signal of the current frame obtained by the time-domain preprocessing into: performing high-pass filtering on the left and right channel time-domain signals of the current frame, respectively, to obtain. The left channel time domain signal of the current frame obtained by preprocessing is denoted as x _{L_HP} (n), and the right channel time domain signal of the current frame obtained by preprocessing is denoted as x _{R_HP} (n). n is the sequence number of sampling points, n=0, 1, . . . , and H−1. The left channel time domain signal of the current frame obtained by time domain preprocessing and the right channel time domain signal of the current frame obtained by time domain preprocessing are the left and right channels of the current frame obtained by time domain preprocessing. It can be called a time domain signal. An Infinite Impulse Response (IIR) digital filter with a cutoff frequency of 20 Hz (Hz) may be used during the high-pass filtering process, or other types of filters may be used.

For example, when the sampling rate of a stereo signal is 16kHz, the corresponding transfer function of a high-pass filter with a cutoff frequency of 20Hz can be:

b0 = 0.994461788958195, b1 = -1.988923577916390, b2 = _{0.994461788958195} , a1 _{= 1.98892905899653} , _{a2 =} -0.988954249933127, and z represents the transform coefficient of the Z transform. The corresponding time domain filters are:
x _{L_HP} (n) = b ₀ x _L (n) + b ₁ x _L (n-1) + b ₂ x _L (n-2) - a ₁ x _{L_HP} (n-1) - a ₂ x _{L_HP} (n-2) (18)

402: Perform time domain analysis on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing.

Specifically, time domain analysis may include transient detection and the like. Transient detection separately performs energy detection on the left channel time domain signal and right channel time domain signal of the current frame obtained by preprocessing to detect whether an energy burst occurs in the current frame. It can be

For example, the energy E _{cur_L} of the left channel time domain signal of the current frame obtained by preprocessing is calculated. Transient detection uses the energy E _{pre_L} of the left channel time domain signal of the frame before the current frame obtained by preprocessing to obtain the transient detection result of the left channel time domain signal of the current frame obtained by preprocessing. It is based on the absolute value of the difference between the energy _{Ecur_L} of the left channel time domain signal of the current frame obtained by preprocessing. Transient detection can be performed on the right channel time domain signal of the current frame obtained by preprocessing using the same method.

The time domain analysis may include other prior art time domain analyzes in addition to transient detection. For example, time domain analysis may include time domain Inter-channel Time Difference (ITD) parameter determination, time domain delay matching processing, and band extension preprocessing.

403: Perform a time-frequency transform on the left channel time domain signal and the right channel time domain signal obtained by the time domain preprocessing to obtain a left channel frequency domain signal and a right channel frequency domain signal.

Specifically, a discrete Fourier transform may be performed on the left channel time domain signal obtained by the time domain preprocessing to obtain the left channel frequency domain signal, and to obtain the right channel frequency domain signal: A discrete Fourier transform is performed on the right channel time domain signal obtained by the time domain preprocessing.

To overcome the problem of spectral aliasing, a convolution-add method may be used to process the Discrete Fourier Transform between two consecutive times, in some cases adding zeros to the input signal of the Discrete Fourier Transform. obtain.

The discrete Fourier transform may be performed once per frame, or each frame of the signal may be divided into P subframes, where P is a positive integer greater than or equal to 2, and the discrete Fourier transform may be performed once per subframe. is performed once every

For example, the Discrete Fourier Transform is performed once for the current frame, the left channel frequency domain signal of the current frame on which the Discrete Fourier Transform is performed is denoted as L(k), and the current frame on which the Discrete Fourier Transform is performed is denoted R(k). k represents the frequency bin index number, k=0, 1, . The current frame being read contains L frequency bins.

In another example, the current frame is partitioned into P subframes, where P is a positive integer greater than or equal to 2. The left channel frequency domain signal of the subframe where the discrete Fourier transform is performed with index number _i is denoted by Li(k), and the right channel frequency of the subframe where the discrete Fourier transform is performed with index number i The area signal is denoted R _i (k). i represents the subframe index number, i = 0, 1, ..., P-1, k represents the frequency bin index number, k = 0, 1, ..., L-1, and L is , represents the frame length of each subframe on which the discrete Fourier transform is performed, in other words, each subframe on which the discrete Fourier transform is performed contains L frequency bins.

404: Determine ITD parameters and encode the determined ITD parameters.

Specifically, there are multiple methods for determining the ITD parameters. The ITD parameters may be determined only in the frequency domain, only in the time domain, or determined in the time-frequency domain. This is not a limitation in this application.

および

が計算される。 ITD parameters can be extracted in the time domain using cross-correlation coefficients. For example, in the range _0≤i≤Tmax ,

and

is calculated.

の場合、ITDパラメータ値は、max（c_n（i））に対応するインデックス番号の反対の数である。

の場合、ITDパラメータ値は、max（c_p（i））に対応するインデックス番号である。iは、相互相関係数を計算するためのインデックス番号を表し、jは、サンプリング点のインデックス番号を表し、T_maxは、異なるサンプリングレートにおけるITDパラメータ値の最大値に対応し、Hは、現在のフレームのフレーム長を表す。

, the ITD parameter value is the opposite number of the index number corresponding to max(c _n (i)).

, the ITD parameter value is the index number corresponding to max(c _p (i)). i represents the index number for calculating the cross-correlation coefficient, j represents the index number of the sampling point, T _max corresponds to the maximum value of the ITD parameter values at different sampling rates, H is the current represents the frame length of the frame.

The ITD parameters may alternatively be determined in the frequency domain based on the left channel frequency domain signal and the right channel frequency domain signal. For example, time-domain A signal may be transformed into a frequency domain signal.

In this embodiment of the present application, the left channel frequency domain signal of the subframe whose index number is i and the discrete Fourier transform is performed is denoted as L _i (k), where k=0, 1, . . . , L/ 2−1, the index number is i, and the right channel frequency domain signal of the subframe under transformation is denoted R _i (k), where k=0, 1, . . . , L/2−1 and i=0, 1, . . . , P−1. The frequency-domain cross -correlation coefficient of the subframe with index number _i is calculated according to XCORRi(k)=Li(k) R ^* _i (k ₎ , where R ^* _i (k) is the transform performed. represents the conjugate of the right channel frequency-domain signal for the i-th subframe where

であることを得るために、L／2－T_max≦n≦L／2＋T_maxの範囲でxcorr_i（n）の最大値が探索される。 The frequency domain cross-correlation coefficients are transformed to the time domain xcorr _i (n), where n=0, 1, .

The maximum value of xcorr _i (n) is searched for in the range L/2-T _max ≤ n ≤ L/2 + T _max to obtain that .

に従って振幅値がさらに計算されてもよく、ITDパラメータ値は

であり、具体的には、ITDパラメータ値は、最大振幅値に対応するインデックス番号である。 In addition, in the search range −T _max ≤ j ≤ T _max , based on the left channel frequency domain signal and the right channel frequency domain signal of the subframe where the index number is i and the DFT transformation is performed

Amplitude values may be further calculated according to and the ITD parameter value is

and specifically, the ITD parameter value is the index number corresponding to the maximum amplitude value.

After the ITD parameters are determined, the ITD parameters may be encoded to obtain the encoding parameters, which are written into the stereo encoded bitstream.

405: Perform time shift adjustment on the left frequency domain signal and the right channel frequency domain signal based on the ITD parameters.

Specifically, time shift adjustments may be made to the left channel frequency domain signal and the right channel frequency domain signal according to any technique. This is not a limitation in this embodiment of the application.

For example, the current frame of the signal is divided into P subframes, where P is a positive integer greater than or equal to 2. The left channel frequency domain signal obtained by time shift adjustment of the subframe with index number i may be denoted as L' _i (k), where k = 0, 1, ..., L/2-1. , and the right channel frequency domain signal obtained by time shift adjustment of the subframe with index number i may be denoted as R′ _i (k), where k represents the frequency bin index number and k= 0, 1, . . . , L/2−1, i represents the subframe index number, and i=0, 1, .

T _i represents the ITD parameter value of the subframe with index number i, L represents the length of the subframe over which the discrete Fourier transform is performed, L _i (k) is the index number i, R i (k) represents the left channel frequency domain signal of the ith subframe in which the transform is performed, R _i (k) represents the right channel frequency domain signal of the subframe in which the index number is i, where i is Represents the subframe index number, i=0, 1, . . . , P−1.

406: Calculate another frequency domain stereo parameter based on the left channel frequency domain signal and the right channel frequency domain signal obtained by the time shift adjustment, and encode the other frequency domain stereo parameter.

Specifically, the other frequency-domain stereo parameters are Inter-channel Phase Difference (IPD) parameters, and/or Inter-channel Level Difference (ILD) parameters, and/or It may include, but is not limited to, sub-band side gains and the like. ILD may also be referred to as inter-channel amplitude difference.

After obtaining the other frequency-domain stereo parameters by calculation, the other frequency-domain stereo parameters may be coded to obtain the coding parameters, and the coding parameters are written into the stereo-coded bitstream.

407: Determine M subbands that satisfy a preset condition from the N subbands included in the frequency domain signal of the current frame.

Specifically, the frequency domain signal obtained by the time shift adjustment of the current frame is divided into subbands. For example, the frequency domain signal of the current frame is divided into N subbands (where N is a positive integer greater than or equal to 2), and the frequency bin contained in the subband with subband index number b is k ∈ [ band_limits(b), band_limits(b+1)−1], where band_limits(b) represents the minimum index number of frequency bins included in the subband whose subband index number is b, and band_limits(b+1) represents the subband It represents the minimum index number of the frequency bins included in the subband whose band index number is b+1. According to a preset condition, M subbands that satisfy the preset condition are determined among the N subbands.

For example, the preset condition is that the subband index number is less than or equal to the preset maximum subband index number, i.e., b≤res_cod_band_max, where res_cod_band_max represents the preset maximum subband index number; can be

Alternatively, the preset condition is that the subband index number is less than or equal to the maximum preset subband index number and greater than or equal to the minimum preset subband index number, i.e. res_cod_band_min≤b≤res_cod_band_max, res_cod_band_max is , res_cod_band_min represents the preset maximum subband index number, and res_cod_band_min represents the preset minimum subband index number.

Furthermore, for wideband stereo signals, different preset conditions may be set based on different coding rates. For example, when the coding rate is 26 kbps, the preset condition is subband index number b≦5, in other words, the preset maximum subband index number is 5. When the coding rate is 44 kbps, the preset condition is subband index number b≦6, in other words, the preset maximum subband index number is 6. When the coding rate is 56 kbps, the preset condition is subband index number b≦7, in other words, the preset maximum subband index number is 7.

For example, if the preset condition is subband index number b≤4, the 5 subbands with index numbers 0 to 4 are the subbands satisfying the preset condition among the N subbands of the current frame. can be determined as a band.

In addition, if the current frame of the signal is divided into P subframes (P is a positive integer greater than or equal to 2), each subframe obtained by time shift adjustment is divided into subbands. For example, a subframe with an index number of i (i = 0, 1, ..., P-1) is divided into N subbands, and subbands with an index number of b in the subframe with an index number of i The frequency bins included in the band are k _i ∈ [band_limits(b), band_limits(b+1)−1], where band_limits(b) is the subband with index number b in the subframe with index number i. and band_limits(b+1) represents the minimum index number of frequency bins included in the subband with index number b+1 in the subframe with index number i.

According to a preset condition, M subbands that satisfy the preset condition are determined from among the N subbands included in each frame.

The preset condition may be that the subband index number is greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number, ie, res_cod_band_min≤b≤res_cod_band_max.

For example, if the preset condition is 4≤b≤8, the 5 subbands with index numbers 4 to 8 satisfy the preset condition among the N subbands in each subframe. is determined as

408: Based on the left channel frequency domain signal and the right channel frequency domain signal obtained by the time shift adjustment, calculate the sub-band downmix signal and the residual signal that satisfy the preset condition.

Specifically, the method for calculating the subband downmix signal and residual signal satisfying the preset condition is that the current frame has P subframes (P is a positive integer greater than or equal to 2). It will be described using an example of partitioning (eg, the current frame may be partitioned into 2 subframes or 4 subframes).

For example, if the preset condition is that the subband index number b is less than or equal to 5, the downmix signal and residual signal of the subbands with index numbers from 0 to 5 in each subframe are calculated. .

RES_i’（k）＝RES_i（k）－g＿ILD_i・DMX_i（k）（21）

β＝arctan（sin（IPD_i（b）），cos（IPD_i（b））＋2・c）（24）、および

A downmix signal of a subband whose index number is b (b≤5) in the subframe whose index number is i is denoted as DMX _i (k), and the index number in the subframe whose index number is i is The residual signal of subband b is denoted RES _i '(k), where DMX _i (k) and RES _i '(k) satisfy the following equations.

RES _i '(k) = RES _i (k) - g_ILD _i DMX _i (k) (21)

β = arctan(sin(IPD _i (b)), cos (IPD _i (b)) + 2 c) (24), and

IPD _i (b) represents the IPD parameters of the subband whose index number is b in the subframe whose index number is i, and g_ILD _i is the index number of b in the subframe whose index number is i. represents the side gain of a subband, L' _i (k) is the left channel frequency domain signal obtained by time shift adjustment of the subband with index number b in the subframe with index number i , R′ _i (k) represents the right channel frequency domain signal obtained by time shift adjustment of the subband with index number b in the subframe with index number i, and L′ ' _i (k) represents the left channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number _i , and R''i (k) represents the right channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number i, where i is the subframe index; number, i=0, 1, . represents the minimum index number of frequency bins included in the subband with index number b in the subframe with index number i, and band_limits(b+1) is the index number in the subframe with index number i. represents the lowest index number of the frequency bin contained in the subband where is b+1.

In another example, the downmix signal DMX _i (k) for the subband with index number b in the subframe with index number i may alternatively be calculated according to the following method.
DMX _i (k) = [L''(k) + R''(k)] c(26), and

L'' _i (k) represents the left channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number i, and R'' _i (k) represents the right channel frequency domain signal obtained by adjusting multiple stereo parameters of the subband with index number b in the subframe with index number i, i is the sub represents the frame index number, i=0, 1, . b) represents the minimum index number of frequency bins included in the subband whose subband index number is b, and band_limits(b+1) is the subband whose index number is b+1 in the subframe whose index number is i. represents the lowest index number of the frequency bins contained in . The method for calculating downmix signal energy and residual signal energy is not limited in this embodiment of the application.

409: Determining residual signal coding parameters based on the downmix signal energy and the residual signal energy of the sub-bands satisfying a preset condition.

410: Determine whether the residual signal of each of the M subbands of the current frame needs to be coded based on the residual signal coding parameters. If it is determined that the residual signal needs to be encoded, 412 is performed. If it is determined that the residual signal does not need to be encoded, 411 is performed.

411: Encoding the downmix signal of each of the M subbands of the current frame based on the residual signal encoding parameters. In this case the residual signal does not need to be coded.

412: Encode the downmix signal and the residual signal for each of the M subbands of the current frame based on the residual signal coding parameters.

See the related description of method 300 for specific implementations of steps 409 through 411 . For the sake of brevity, the details are not repeated here.

In method 300 , the encoder side divides the current frame into P subframes, where P is a positive integer greater than or equal to 2, and divides the spectral coefficients of each of the P subframes into N subbands. , and a downmix signal of M subbands (the M subbands are part of at least the N subbands) in each subframe, where the residual signal coding parameter satisfies a preset condition energy and residual signal energy, then in method 300 the residual signal energy res_cod_NRG_S[b] for the subband with index number b in the current frame is is the sum of the residual signal energy of the subband with index number b in the current frame, and the downmix signal energy res_cod_NRG_M[b] of the subband with index number b in the current frame is the sum of all P subframes is the sum of the downmix signal energies of the subband whose index number is b.

For example, the current frame is divided into two subframes and the spectral coefficients of each of the two subframes are divided into N subbands. Therefore, in method 300, the downmix signal energy res_cod_NRG_M[b] of the subband with index number b in the current frame is equal to the downmix signal energy of the subband with index number b in subframe 1. The sum of the downmix signal energy of the subband with index number b in frame 2, and the residual signal energy res_cod_NRG_S[b] of the subband with index number b in the current frame is is the sum of the residual signal energy of the subband with index number b in subframe 2 and the residual signal energy of the subband with index number b in subframe 2 .

The stereo signal encoding method according to the embodiment of the present application has been described in detail above with reference to FIGS. 1 to 7. FIG. The stereo signal encoding device in the embodiment of the present application will be described below with reference to FIGS. 8 and 9. FIG. It should be understood that both the devices in FIGS. 8 and 9 correspond to the stereo signal encoding method in the embodiments of the present application. In addition, any of the devices in Figures 8 and 9 can perform the stereo signal encoding method in the embodiments of the present application. For the sake of brevity, repetitive descriptions are omitted where appropriate.

FIG. 8 is a schematic block diagram of a stereo signal encoding device according to one embodiment of the present application. Apparatus 500 of FIG.
A first determination configured to determine a residual signal coding parameter for the current frame of the stereo signal based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame. A module 501, wherein the residual signal coding parameter of the current frame is used to indicate whether to code the residual signal of M subbands, where M subbands are N subbands. a first determining module 501 that is at least part of a band, where N is a positive integer greater than 1, M≦N, where M is a positive integer;
a second decision module 502, configured to decide whether to code the residual signal of the M subbands of the current frame based on the residual signal coding parameters of the current frame; include.

In this application, the residual signal coding parameters are determined based on the downmix signal energy and the residual signal energy of M subbands within the N subbands that satisfy a preset bandwidth range. , whether to code the residual signal of each of the M subbands is determined based on the residual signal coding parameters. This avoids encoding only the downmix signal when the encoding rate is relatively low. Alternatively, whether to code all residual signals in subbands satisfying a preset bandwidth range is determined based on residual signal coding parameters. Therefore, the spatial sensation and sound image stability of the decoded stereo signal are improved, and at the same time, the high frequency distortion of the decoded stereo signal can be reduced as much as possible, thereby improving the coding quality.

Optionally, in one implementation, the M subbands are the M subbands whose subband index numbers are less than or equal to a preset maximum subband index number in the N subbands.

Optionally, in one embodiment, the M subbands have a subband index number greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number in the N subbands. are M subbands.

Optionally, in one implementation, the second determining module 502 compares the residual signal coding parameter to a preset first threshold, if the first threshold is greater than 0 and less than 1.0, determining not to encode the residual signal for each of the M subbands if the residual signal coding parameter is less than or equal to the first threshold, or the residual signal coding parameter is greater than the first threshold; If so, determine to encode the residual signal of each of the M subbands.

Optionally, in one implementation, the first determining module 501 determines the residual signal coding parameters based on the downmix signal energy, the residual signal energy and the side gains for each of the M subbands. is further configured as

Optionally, in one implementation, the first determining module 501 determines the first parameter based on the downmix signal energy, the residual signal energy and the side gain for each of the M subbands; A parameter of 1 indicates a value relationship between the downmix signal energy and the residual signal energy of each of the M subbands, based on the downmix signal energy and the residual signal energy of each of the M subbands. to determine a second parameter, the second parameter indicating a value relationship between the first energy sum and the second energy sum, the first energy sum being the residual signal energy of the M subbands and the downmix signal energy, the second energy sum is the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency domain signal of the previous frame of the current frame, and the current where the M subbands of the frame have the same subband index numbers as the M subbands of the previous frame, and the long-term smoothing of the first parameter, the second parameter, and the frame before the current frame It is further configured to determine residual signal coding parameters based on the parameters.

Optionally, in one implementation, the first determining module 501 determines M energy parameters based on downmix signal energy, residual signal energy, and side gain for each of the M subbands; The M energy parameters each indicate a value relationship between the downmix signal energy and the residual signal energy for one of the M subbands, and the M energy parameters correspond to the M subbands. It is further configured to determine the energy parameter having the largest value among the M energy parameters with a one-to-one correspondence as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by the first determination module 501 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/(res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g(b) is 4 represents a function of the side gain side_gain[b] of the subband whose subband index number is b.

Optionally, in one implementation, the first determining module 501 determines the first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being , a value relationship between the downmix signal energy and the residual signal energy for each of the M subbands, and based on the downmix signal energy and the residual signal energy for each of the M subbands, a second determining a parameter, the second parameter indicating a value relationship between the first energy sum and the second energy sum, the first energy sum being the residual signal energy of the M subbands and the downmix signal; is the sum of the energies of the current frame, and the second energy sum is the sum of the residual signal energies and the downmix signal energies of the M subbands in the frequency-domain signal of the previous frame of the current frame, and the M subbands have the same subband index numbers as the M subbands of the previous frame, based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame It is further configured to determine residual signal coding parameters.

Optionally, in one implementation, the first determining module 501 determines M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, and determines the M energy The parameters each indicate a value relationship between the downmix signal energy and the residual signal energy of each of the M subbands, the M energy parameters correspond one-to-one with the M subbands, and M It is further configured to determine the energy parameter having the maximum value of the energy parameters as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by the first determination module 501 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] res_cod_NRG_M[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b.

Optionally, in one implementation, the first determination module 501 determines the sum of the M energy parameters as the first parameter res_dmx_ratio ₁ (to be corrected), res_dmx_ratio ₁ being the sum of the M energy parameters of which res_dmx_ratio_max and the downmix signal energy res_cod_NRG_M[b] of each of the M sub-bands, and determining res_dmx_ratio ₂ obtained by the correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, in one implementation, the res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

Optionally, in one implementation, the first determining module 501 bases the first parameter on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands. further configured to determine the

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, in one implementation, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the following equation:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, in one implementation, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the following equation:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by the first determination module 501 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. show.

Optionally, in one embodiment, the residual signal coding parameter of the current frame determined by the first determination module 501 is the long-term smoothing parameter of the current frame, and the long-term smoothing parameter of the current frame is The parameters satisfy the following formula,
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is the same as the preset second is greater than the value of α for greater than or equal to the threshold of , the second threshold is between 0 and 0.6 inclusive, the third threshold is between 2.7 and 3.7 inclusive, or the second parameter is prior to The value of α when the first parameter is greater than the preset fourth threshold, if greater than the preset fifth threshold is is greater than the value of α, the fourth threshold is between 0 and 0.9, the fifth threshold is between 0 and 0.71, or the first parameter is preset above the second threshold The value of α is small and the second parameter is greater than the preset third threshold, the second parameter is greater than or equal to the preset fifth threshold and is less than or equal to the preset third threshold , the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0.71 It is below.

Optionally, in one embodiment, the second determining module 502, when it is determined to encode the M subband residual signals, encodes the M subband downmix signals and the residual signals. It is further configured to encode the downmix signal of the M subbands when it is determined to encode or not to encode the residual signal of the M subbands.

FIG. 9 is a schematic block diagram of a stereo signal encoding device according to one embodiment of the present application. Apparatus 600 of FIG.
a memory 601 configured to store a program;
A processor 602 configured to execute a program stored in a memory 601, wherein when the program in the memory is executed, the processor 602 converts residual signal coding parameters of a current frame of a stereo signal into: determined based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame, wherein the residual signal coding parameters of the current frame encode the residual signals of the M subbands is used to indicate whether the M subbands are at least a portion of the N subbands, N is a positive integer greater than 1, M ≤ N, where N is a positive integer A processor 602, specifically configured to determine whether to encode the residual signal of the M subbands of the current frame based on the residual signal coding parameters.

Optionally, in one implementation, the M subbands are the M subbands whose subband index numbers are less than or equal to a preset maximum subband index number in the N subbands.

Optionally, in one embodiment, the M subbands have a subband index number greater than or equal to a preset minimum subband index number and less than or equal to a preset maximum subband index number in the N subbands. are M subbands.

In one optional implementation, the processor 602 compares the residual signal coding parameter to a preset first threshold, and if the first threshold is greater than 0 and less than 1.0, the residual signal coding If the parameter is less than or equal to the first threshold, determine not to code the residual signal for each of the M subbands; or if the residual signal coding parameter is greater than the first threshold, M It is further configured to determine to encode the residual signal of each of the subbands.

In an optional implementation, the processor 602 is further configured to determine residual signal coding parameters based on downmix signal energy, residual signal energy, and side gains for each of the M subbands. be.

In one optional implementation, processor 602 determines a first parameter based on the downmix signal energy, residual signal energy, and side gain for each of the M subbands, the first parameter being: indicating a value relationship between the downmix signal energy and the residual signal energy for each of the M subbands, a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands; , where the second parameter indicates a value relationship between the first energy sum and the second energy sum, and the first energy sum is the residual signal energy and the downmix signal energy of the M subbands and the second energy sum is the sum of the residual signal energies and the downmix signal energies of M subbands in the frequency domain signal of the previous frame of the current frame, and the M has the same subband index number as the M subbands of the previous frame, and the residual is calculated based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame. It is further configured to determine difference signal encoding parameters.

In one optional implementation, the processor 602 determines M energy parameters based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands; respectively denote the value relationship between the downmix signal energy of one of the M subbands and the residual signal energy, and the M energy parameters correspond one-to-one with the M subbands and determining the energy parameter having the maximum value among the M energy parameters as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by processor 602 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/(res_cod_NRG_S[b] + (1-g(b)) (1-g(b)) res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g(b) is 4 represents a function of the side gain side_gain[b] of the subband whose subband index number is b.

In one optional implementation, the processor 602 determines the first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the first parameter being the indicating a value relationship between the downmix signal energy and the residual signal energy for each of the bands, determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands; A second parameter indicates a value relationship between the first energy sum and the second energy sum, the first energy sum being the sum of the residual signal energy and the downmix signal energy of the M subbands. , the second energy sum is the sum of the residual signal energy and the downmix signal energy of the M subbands in the frequency domain signal of the previous frame of the current frame, and the M subbands of the current frame are Residual signal encoding based on the first parameter, the second parameter, and the long-term smoothing parameter of the frame before the current frame, with the same subband index number as the M subbands of the previous frame It is further configured to determine parameters.

In one optional implementation, the processor 602 determines M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, the M energy parameters being equal to the M and the residual signal energy of each of the subbands of the M energy parameters corresponding one-to-one with the M subbands, and the M energy parameters of It is further configured to determine the energy parameter having the maximum value thereof as the first parameter.

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by processor 602 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b, b is greater than or equal to 0 and less than or equal to the preset maximum subband index number, res_cod_NRG_S[b] res_cod_NRG_M[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b.

In one optional implementation, the processor 602 determines the sum of the M energy parameters as the first parameter res_dmx_ratio ₁ (to be corrected), where res_dmx_ratio ₁ is the maximum value of the M energy parameters. It is further configured to correct based on the res_dmx_ratio_max and the downmix signal energy res_cod_NRG_M[b] of each of the M subbands and determine a res_dmx_ratio ₂ obtained by the correction.

For example, the encoder side corrects res_dmx_ratio ₁ according to the following formula, where M=5,
res_dmx_ratio ₂ obtained by correction satisfies the following equation.

Optionally, in one implementation, the res_dmx_ratio ₂ obtained by correction may be further corrected.

For example, res_dmx_ratio ₃ finally obtained by correction satisfies the following formula,
_{res_dmx_ratio3} = pow( _{res_dmx_ratio2} , 1.2)
In the formula, the pow() function represents an exponential function, and pow( _{res_dmx_ratio2} , _1.2 ) represents res_dmx_ratio2 raised to the power of 1.2.

Optionally, in one implementation, the processor 602 determines the first parameter based on the sum of the residual signal energies of the M subbands and the sum of the downmix signal energies of the M subbands. further configured to

Specifically, the encoder side separately determines the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands and the sum res_nrg_all_curr of the residual signal energies of the M subbands, and based on dmx_nrg_all_curr and res_nrg_all_curr Determine the first parameter.

式中、res＿cod＿NRG＿M＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドのダウンミックス信号エネルギーを表し、γ₁は、平滑化係数を表し、γ₁は、0以上1以下の実数であり、例えば、γ₁＝0．1である。 Optionally, in one implementation, the sum dmx_nrg_all_curr of the downmix signal energies of the M subbands satisfies the following equation:

where _{res_cod_NRG_M_prev} [b] represents the downmix signal energy of the subband with subband index number b in the frame previous to the current frame, γ1 represents the smoothing factor, and γ1 is ₀ It is a real number greater than or equal to 1 and less than or equal to 1, for example, γ ₁ =0.1.

式中、res＿cod＿NRG＿S＿prev［b］は、現在のフレームの前のフレームにおけるサブバンドインデックス番号がbであるサブバンドの残差信号エネルギーを表し、γ₂は、平滑化係数を表し、γ₂は、0以上1以下の実数であり、例えば、γ₂＝0．1である。 Optionally, in one implementation, the sum res_nrg_all_curr of the residual signal energies of the M subbands satisfies the following equation:

where res_cod_NRG_S_prev[ _b ] represents the residual signal energy of the subband with subband index number _b in the frame previous to the current frame, γ2 represents the smoothing factor, and γ2 is 0 It is a real number greater than or equal to 1 and less than or equal to 1. For example, γ ₂ =0.1.

The encoder side determines the first parameter res_dmx_ratio based on dmx_nrg_all_curr and res_nrg_all_curr.

For example, the first parameter res_dmx_ratio finally determined by the encoder side satisfies the following equation.
res_dmx_ratio = res_nrg_all_curr/dmx_nrg_all_curr

Optionally, in one embodiment, the energy parameter of the subband with subband index number b among the M energy parameters determined by processor 602 satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband whose subband index number is b among the M energy parameters, b is greater than or equal to 0, and is the preset maximum subband index number. where res_cod_NRG_S[b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. show.

Optionally, in one implementation , if the first parameter is less than a preset second threshold and the second parameter is greater than a preset third threshold, the processor 602 The residual signal coding parameter determined by is the long-term smoothing parameter of the current frame, and the long-term smoothing parameter of the current frame satisfies the following equation,
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the frame before the current frame, 0<α<1 and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is the same as the preset second is greater than the value of α for greater than or equal to the threshold of , the second threshold is between 0 and 0.6 inclusive, the third threshold is between 2.7 and 3.7 inclusive, or the second parameter is prior to The value of α when the first parameter is greater than the preset fourth threshold, if greater than the preset fifth threshold is is greater than the value of α, the fourth threshold is between 0 and 0.9, the fifth threshold is between 0 and 0.71, or the first parameter is preset above the second threshold The value of α is small and the second parameter is greater than the preset third threshold, the second parameter is greater than or equal to the preset fifth threshold and is less than or equal to the preset third threshold , the second threshold is 0 or more and 0.6 or less, the third threshold is 2.7 or more and 3.7 or less, and the fifth threshold is 0 or more and 0.71 It is below.

Optionally, in one embodiment, the processor 602 encodes the M subband downmix and residual signals when it is determined to encode the M subband residual signals. , or is further configured to encode the downmix signal of the M subbands when it is determined not to encode the residual signal of the M subbands.

The present application further provides chips. The chip includes a processor and communication interface. The communication interface is configured to communicate with an external device, and the processor is configured to perform the stereo signal encoding method in the embodiments of the present application.

Optionally, in one implementation, the chip may further include memory. The memory stores instructions and the processor is configured to execute the instructions stored in the memory. When the instructions are executed, the processor is configured to perform the stereo signal encoding method in the embodiments of the present application.

Optionally, in one embodiment, the chip is incorporated into terminal equipment or network equipment.

The present application provides a computer-readable storage medium. The computer-readable storage medium stores program code to be executed by the device. The program code includes instructions for performing the stereo signal encoding method in the embodiments of the present application.

The processors referred to in the embodiments of the present invention may be Central Processing Units (CPUs), or other general purpose processors, Digital Signal Processors (DSPs), application specific integrated circuits. (Application Specific Integrated Circuit, ASIC), Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. . A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, and so on.

It will be appreciated that the memory referred to in embodiments of the present invention may be volatile memory or non-volatile memory, and may include volatile and non-volatile memory. Nonvolatile memory includes Read-Only Memory (ROM), Programmable Read-Only Memory (Programmable ROM, PROM), Erasable Programmable Read-Only Memory (Erasable PROM, EPROM), Electrically Erasable Programmable Read-Only Memory (Electrically EPROM, EEPROM), or flash memory. Volatile memory can be random access memory (RAM), used as an external cache. By way of example and not limitation, many forms of RAM such as Static Random Access Memory (Static RAM, SRAM), Dynamic Random Access Memory (Dynamic RAM, DRAM), Synchronous Dynamic Random Access Memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), sync Link Dynamic Random Access Memory (Synchlink DRAM, SLDRAM), Direct Rambus Random Access Memory (Direct Rambus RAM, DR RAM) may be used.

Note that if the processor is a general-purpose processor, DSP, ASIC, FPGA, or other programmable logic device, discrete gates, transistor logic circuits, or discrete hardware components, memory (storage module) is integrated into the processor. sea bream.

Note that the memory described herein includes, but is not limited to, these memories and any other suitable types of memory.

Those skilled in the art will understand that each unit and algorithm step can be realized by electronic hardware or by a combination of computer software and electronic hardware in combination with the examples described in the embodiments disclosed herein. would do. Whether the function is performed by hardware or by software depends on the particular application and design constraints of the technical solution. Skilled artisans may implement the described functionality using varying methods for each particular application, but such implementations should not be considered beyond the scope of this application.

For convenience of explanation, the detailed operation processes of the aforementioned systems, devices, and units shall refer to the corresponding processes in the aforementioned method embodiments, and are not described in detail herein. This will be clearly understood by those skilled in the art.

It should be appreciated that in some of the embodiments provided in this application, the disclosed systems, devices, and methods may be implemented in other ways. For example, the described apparatus embodiment is merely exemplary. For example, the division into units is merely a logical functional division, and other divisions are possible in actual implementation. For example, multiple units or components may be combined or integrated into other systems, and some features may be ignored or not implemented. Additionally, the mutual couplings or direct couplings or communication connections shown or described may be implemented using some interfaces. Indirect couplings or communicative connections between devices or units may be realized in electronic, mechanical, or other form.

Units described as separate parts may or may not be physically separate and parts illustrated as units may or may not be physical units and may or may not be located together. may be distributed over multiple network units. Part or all of the units may be selected based on actual requirements to achieve the purpose of each embodiment's solution.

In addition, the functional units in the embodiments of the present application may be integrated into one processing unit, or each of the units may exist physically independently, or two or more units may be combined into one integrated into the unit.

When each function is implemented in the form of software functional units and sold or used as an independent product, the functions can be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the present application may be realized in the form of software products essentially or part of which contributes to the prior art, or part of the technical solutions. The computer software product is stored in a storage medium and instructs a computing device (which may be a personal computer, server, network appliance, etc.) to perform all or part of the method steps described in the embodiments of the present application. Including some commands to command. The aforementioned storage medium may store the program code, such as a USB flash drive, removable hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk. including any medium capable of

The above descriptions are only specific embodiments of the present application and are not intended to limit the protection scope of the present application. Any variation or replacement readily figured out by a person skilled in the art within the technical scope disclosed in the present application shall fall within the protection scope of the present application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

110 Encoding Components
120 decoding components
130 mobile terminals
131 collection components
132 channel coding components
140 mobile terminals
141 Audio Playback Components
142 channel decoding component
150 network elements
151 channel decoding component
152 channel coding components
300 stereo signal encoding method
500 devices
501 First Decision Module
502 second decision module
600 devices
601 memory
602 processor

Claims (22)

A stereo signal encoding method comprising:
determining a residual signal coding parameter for a current frame of a stereo signal based on the downmix signal energy and the residual signal energy of each of the M subbands of the current frame; The residual signal coding parameter of a frame is used to indicate whether to code residual signals of the M subbands, the M subbands being at least part of the N subbands. where N is a positive integer greater than 1, M≤N, where M is a positive integer;
determining whether to code the residual signals of the M subbands of the current frame based on the residual signal coding parameters of the current frame. Method. The step of determining whether to encode the residual signal of the M subbands based on the residual signal coding parameters of the current frame comprises:
comparing the residual signal coding parameter of the current frame with a preset first threshold, wherein the first threshold is greater than 0 and less than 1.0;
determining not to encode the residual signal of the M subbands if the residual signal coding parameter for the current frame is less than or equal to the first threshold; or and determining to encode the residual signals of the M subbands if a parameter is greater than the first threshold. The step of determining residual signal coding parameters for the current frame based on downmix signal energy and residual signal energy for each of the M subbands,
determining the residual signal coding parameters for the current frame based on the downmix signal energy, the residual signal energy, and side gains for each of the M subbands; The method described in 2. determining the residual signal coding parameters for the current frame based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands;
determining a first parameter based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands, the first parameter comprising the M indicating a value relationship between the downmix signal energy and the residual signal energy for each of the subbands of
determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the second parameter being a sum of a first energy and a second energy sum, wherein the first energy sum is the sum of residual signal energy and downmix signal energy of the M subbands, and the second energy sum is the current a sum of residual signal energy and downmix signal energy of M subbands in a frequency domain signal of a frame previous to the frame, wherein the M subbands of the current frame are the M subbands of the previous frame; having the same subband index number as the subbands of
determining the residual signal coding parameter for the current frame based on the first parameter, the second parameter, and a long-term smoothing parameter for the previous frame of the current frame; 4. The method of claim 3. The step of determining a first parameter based on the downmix signal energy, the residual signal energy, and the side gain for each of the M subbands comprises:
determining M energy parameters based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands, wherein the M energy parameters are the respectively denoting the value relationship between the downmix signal energy and the residual signal energy of each of M subbands, wherein the M energy parameters correspond one-to-one with the M subbands; , step and
5. The method of claim 4, comprising determining an energy parameter having the largest value among the M energy parameters as the first parameter. Among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies the following equation,
res_dmx_ratio[b]=res_cod_NRG_S[b]/(res_cod_NRG_S[b]+(1-g(b))・(1-g(b))res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband with subband index number b, b is greater than or equal to 0 and less than or equal to a preset maximum subband index number, and res_cod_NRG_S[ b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g( 6. The method of claim 5, wherein b) represents a function of the side gain side_gain[b] of the subband with subband index number b. The step of determining residual signal coding parameters for the current frame based on downmix signal energy and residual signal energy for each of the M subbands,
determining a first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, wherein the first parameter is for each of the M subbands; indicating a value relationship between the downmix signal energy and the residual signal energy of
determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the second parameter being a sum of a first energy and a second energy sum, wherein the first energy sum is the sum of residual signal energy and downmix signal energy of the M subbands, and the second energy sum is the current a sum of residual signal energy and downmix signal energy of M subbands in a frequency domain signal of a frame previous to the frame, wherein the M subbands of the current frame are the M subbands of the previous frame; having the same subband index number as the subbands of
determining the residual signal coding parameter for the current frame based on the first parameter, the second parameter, and a long-term smoothing parameter for the previous frame of the current frame; 3. A method according to claim 1 or 2. The step of determining a first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands comprises:
determining M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, wherein the M energy parameters are respectively representing the value relationship between the downmix signal energy and the residual signal energy of each of the M energy parameters corresponding one-to-one with the M subbands;
8. The method of claim 7, comprising determining the energy parameter having the largest value among the M energy parameters as the first parameter. Among the M energy parameters, the energy parameter of the subband whose subband index number is b satisfies the following equation,
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband with subband index number b, b is greater than or equal to 0 and less than or equal to a preset maximum subband index number, and res_cod_NRG_S[ b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. The method described in 8. the residual signal coding parameter of the current frame is a long-term smoothing parameter of the current frame, wherein the long-term smoothing parameter of the current frame satisfies the following equation:
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, and res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the previous frame of the current frame. , 0<α<1, and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is greater than the value of α when it is equal to or greater than the set second threshold, the second threshold is 0 or more and 0.6 or less, and the third threshold is 2.7 or more and 3.7 or less, or or if said second parameter is greater than a preset fifth threshold, the value of α if said first parameter is greater than a preset fourth threshold, then said first parameter is greater than said greater than the value of α when less than or equal to a pre-set fourth threshold, wherein said fourth threshold is 0 or more and 0.9 or less and said fifth threshold is 0 or more and 0.71 or less, or If the second parameter is greater than or equal to a preset fifth threshold and less than or equal to a preset third threshold, then the value of α is equal to or greater than the preset second threshold for the first parameter. is less than the value of α when the second parameter is greater than the preset third threshold, the second threshold is 0 or more and 0.6 or less, and the third threshold is 10. A method according to any one of claims 4 to 9, wherein the fifth threshold is between 0 and 0.71 and is between 2.7 and 3.7. encoding downmix signals and the residual signals for the M subbands when determined to encode the residual signals for the M subbands; or 11. The method of any one of claims 1 to 10, further comprising: encoding downmix signals of the M subbands when it is determined not to encode the residual signal. A stereo signal encoding device,
a memory configured to store a program;
A processor configured to execute the program stored in the memory, wherein when the program in the memory is executed, the processor outputs residual signal coding parameters of a current frame of a stereo signal based on the downmix signal energy and residual signal energy of each of the M subbands of the current frame, wherein the residual signal coding parameters of the current frame are the M subbands wherein the M subbands are at least part of N subbands, N is a positive integer greater than 1, and M≤N , M is a positive integer, and determines whether to code the residual signal of the M subbands of the current frame based on the residual signal coding parameters of the current frame. A stereo signal coding apparatus comprising a processor configured to: . The processor
comparing the residual signal coding parameter to a preset first threshold, wherein the first threshold is greater than 0 and less than 1.0;
if the residual signal coding parameter of the current frame is less than or equal to the first threshold, determine not to code the residual signal of the M subbands; or 13. The apparatus of claim 12, further configured to determine to encode the residual signal of the M subbands if a residual signal coding parameter is greater than the first threshold. The processor
further configured to determine the residual signal coding parameters for the current frame based on the downmix signal energy, the residual signal energy, and side gains for each of the M subbands. 14. Apparatus according to paragraph 12 or 13. The processor
determining a first parameter based on the downmix signal energy, the residual signal energy, and the side gain for each of the M subbands, wherein the first parameter is for each of the M subbands; indicating a value relationship between each of said downmix signal energy and said residual signal energy;
determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the second parameter being a ratio of a first energy sum and a second energy sum; wherein the first energy sum is the sum of residual signal energies and downmix signal energies of the M subbands, and the second energy sum is the a sum of residual signal energy and downmix signal energy of M subbands in a frequency domain signal of a frame, wherein the M subbands of the current frame are the M subbands of the previous frame; have the same subband index number,
further configured to determine the residual signal coding parameter for the current frame based on the first parameter, the second parameter, and a long-term smoothing parameter for the previous frame of the current frame. 15. The apparatus of claim 14, wherein The processor
determining M energy parameters based on the downmix signal energy, the residual signal energy, and the side gains for each of the M subbands, the M energy parameters being equal to the M subbands; each representing the value relationship between the downmix signal energy and the residual signal energy for each of the bands, the M energy parameters corresponding one-to-one with the M sub-bands;
16. The apparatus of claim 15, further configured to determine an energy parameter having a maximum value among said M energy parameters as said first parameter. Among the M energy parameters determined by the processor, the energy parameter of the subband whose subband index number is b satisfies the following equation:
res_dmx_ratio[b]=res_cod_NRG_S[b]/(res_cod_NRG_S[b]+(1-g(b))・(1-g(b))res_cod_NRG_M[b]+1)
where res_dmx_ratio[b] represents the energy parameter of the subband with subband index number b, b is greater than or equal to 0 and less than or equal to a preset maximum subband index number, and res_cod_NRG_S[ b] represents the residual signal energy of the subband with subband index number b, res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b, and g( 17. The apparatus of claim 16, wherein b) represents a function of side gain side_gain[b] for the subband with subband index number b. The processor
determining a first parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, wherein the first parameter is the downmix for each of the M subbands; indicating a value relationship between signal energy and said residual signal energy;
determining a second parameter based on the downmix signal energy and the residual signal energy for each of the M subbands, the second parameter being a ratio of a first energy sum and a second energy sum; wherein the first energy sum is the sum of residual signal energies and downmix signal energies of the M subbands, and the second energy sum is the a sum of residual signal energy and downmix signal energy of M subbands in a frequency domain signal of a frame, wherein the M subbands of the current frame are the M subbands of the previous frame; have the same subband index number,
further configured to determine the residual signal coding parameter for the current frame based on the first parameter, the second parameter, and a long-term smoothing parameter for the previous frame of the current frame. 14. Apparatus according to claim 12 or 13. The processor
determining M energy parameters based on the downmix signal energy and the residual signal energy for each of the M subbands, wherein the M energy parameters are for each of the M subbands; respectively indicating the value relationships between downmix signal energy and the residual signal energy, wherein the M energy parameters correspond one-to-one with the M subbands;
19. The apparatus of claim 18, further configured to determine an energy parameter having a maximum value among said M energy parameters as said first parameter. Among the M energy parameters determined by the processor, the energy parameter of the subband whose subband index number is b satisfies the following equation:
res_dmx_ratio[b] = res_cod_NRG_S[b]/res_cod_NRG_M[b]
where res_dmx_ratio[b] represents the energy parameter of the subband with subband index number b, b is greater than or equal to 0 and less than or equal to a preset maximum subband index number, and res_cod_NRG_S[ b] represents the residual signal energy of the subband with subband index number b, and res_cod_NRG_M[b] represents the downmix signal energy of the subband with subband index number b. 19. Apparatus according to 19. the residual signal coding parameter of the current frame is a long-term smoothing parameter of the current frame, wherein the long-term smoothing parameter of the current frame satisfies the following equation:
res_dmx_ratio_lt = res_dmx_ratio · α + res_dmx_ratio_lt_prev · (1 - α)
where res_dmx_ratio_lt represents the long-term smoothing parameter of the current frame, res_dmx_ratio represents the first parameter, and res_dmx_ratio_lt_prev represents the long-term smoothing parameter of the previous frame of the current frame. , 0<α<1, and
If the second parameter is greater than the preset third threshold, the value of α if the first parameter is less than the preset second threshold is greater than the value of α when it is equal to or greater than the set second threshold, the second threshold is 0 or more and 0.6 or less, and the third threshold is 2.7 or more and 3.7 or less, or or if said second parameter is greater than a preset fifth threshold, the value of α if said first parameter is greater than a preset fourth threshold, then said first parameter is greater than said greater than the value of α when less than or equal to a pre-set fourth threshold, wherein said fourth threshold is 0 or more and 0.9 or less and said fifth threshold is 0 or more and 0.71 or less, or If the second parameter is greater than or equal to a preset fifth threshold and less than or equal to a preset third threshold, then the value of α is equal to or greater than the preset second threshold for the first parameter. is less than the value of α when the second parameter is greater than the preset third threshold, the second threshold is 0 or more and 0.6 or less, and the third threshold is 21. Apparatus according to any one of claims 15 to 20, wherein the fifth threshold is between 0 and 0.71 and is between 2.7 and 3.7. The processor
encoding a downmix signal and the residual signal for the M subbands when it is determined to encode the residual signal for the M subbands; or 22. The apparatus of any one of claims 12-21, further configured to encode downmix signals of the M subbands when it is determined not to encode the residual signal. .

Images